This document is the Accepted Manuscript version of a Published Work that appeared in final form in ACS Nano, copyright © American Chemical Society after peer review and technical editing by the publisher. To access the final edited and published work see: https://dx.doi.org/10.1021/acsnano.9b04898. Neutron Activated 153Sm Sealed in Carbon Nanocapsules for In Vivo Imaging and Tumor Radiotherapy Julie T.-W. Wanga, Rebecca Klippsteina, Markus Martincicb, Elzbieta Pachc, Robert Feldmand, Martin Sefle,f, Yves Micheld, Daniel Askera, Jane K. Sosabowskig, Martin Kalbach, Tatiana Da Rosi, Cécilia Ménard-Moyonj, Alberto Biancoj, Ioanna Kyriakoue, Dimitris Emfietzogloue, Jean- Claude Saccavinid, Belén Ballesterosc,*, Khuloud T. Al-Jamala,* and Gerard Tobiasb,* a Institute of Pharmaceutical Science, King’s College London, London SE1 9NH, UK b Institut de Ciència de Materials de Barcelona (ICMAB-CSIC), Campus UAB, 08193 Bellaterra, Barcelona, Spain c Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC and the Barcelona Institute of Science and Technology, Campus UAB, 08193 Bellaterra, Barcelona, Spain d Cis Bio International Ion Beam Applications SA, Gif sur Yvette 91192, France e Medical Physics Laboratory, University of Ioannina Medical School, Ioannina GR-45110, Greece 1 f Faculty of Nuclear Sciences and Physical Engineering, Czech Technical University in Prague, Prague 11519, Czech Republic g Centre for Molecular Oncology, Barts Cancer Institute, Queen Mary University of London, London EC1M 6BQ, UK h J. Heyrovsky Institute of the Physical Chemistry, Dolejskova 3, 182 23 Prague 8, Czech Republic i INSTM Unit of Trieste, Department of Chemical and Pharmaceutical Sciences, University of Trieste, Via L. Giorgieri 1, 34127 Trieste, Italy j University of Strasbourg, CNRS, Immunology, Immunopathology and Therapeutic Chemistry, UPR 3572, 67000 Strasbourg, France. *Corresponding authors E-mail: gerard.tobias@icmab.es, khuloud.al-jamal@kcl.ac.uk, belen.ballesteros@icn2.cat. 2 ABSTRACT: Radiation therapy along with chemotherapy and surgery remain the main cancer treatments. Radiotherapy can be applied to patients externally (external beam radiotherapy) or internally (brachytherapy and radioisotope therapy). Previously, nanoencapsulation of radioactive crystals within carbon nanotubes, followed by end-closing, resulted in the formation of nanocapsules that allowed ultrasensitive imaging in healthy mice. Herein we report on the preparation of nanocapsules initially sealing ‘cold’ isotopically enriched samarium (152Sm), which can then be activated on demand to their ‘hot’ radioactive form (153Sm) by neutron irradiation. The use of ‘cold’ isotopes avoids the need for radioactive facilities during the preparation of the nanocapsules, reduces radiation exposure to personnel, prevents the generation of nuclear waste and evades the time constraints imposed by the decay of radionuclides. An unprecedented high specific radioactivity is achieved by neutron irradiation (up to 11.37 GBq/mg), making the ‘hot’ nanocapsules useful not only for in vivo imaging but also therapeutically effective against lung cancer metastases after intravenous injection. The high in vivo stability of the radioactive payload, selective toxicity to cancerous tissues and the elegant preparation method offer a new paradigm for application of nanomaterials in radiotherapy. KEYWORDS: cancer therapy, nuclear imaging, nanoencapsulation, filled carbon nanotubes, radiooncology, nanooncology. 3 Advances of nanomedicine in cancer diagnosis and therapy require the production of ‘small’ and ‘smart’ agents that can offer specific targeting, adequate detection sensitivity, efficient therapeutic effects and ideally favorable biocompatibility.1-5 Carbon nanotubes (CNTs) have been exploited as delivery systems for theranostic applications.6-10 Their unique needle-like structure enables efficient cell penetration.11-14 Moreover, a selected biomedically relevant payload can be loaded onto the CNT structure either by chemical modification of the largely available external surface (exohedral), or by filling their interior space (endohedral). Surface engineering is a more straightforward approach and has been widely investigated.15, 16 Although less explored, confinement of imaging and therapeutic agents into CNTs is attracting an increased attention.17, 18 After encapsulation, the external walls remain available to anchor biocompatible and/or targeting ligands. One advantage of the nanoencapsulation approach is the protection that the carbon coating (CNT walls) offers to the encapsulated materials from enzymatic degradation once exposed to biological environment. Only highly defective, oxidized, CNTs are degraded by enzymes.19 Apart from CNTs, a variety of materials are being employed for the formation of nanocapsules that shield a selected payload.20-24 In the vast majority of studies, release of the encapsulated cargo is necessary to achieve a therapeutic effect via the delivery of bioactive agents such as drugs. This is however not the case for nuclear imaging and targeted radiotherapy applications where departure of the encaged imaging and/or therapeutic radionuclides from the nanocapsule is neither required nor desired. From the library of nanocapsules available, the unique properties of carbon nanotubes make them unparalleled for the permanent sealing of radionuclides.25-28 The graphene shells that constitute the walls of the CNTs can perfectly separate the core from the external environment, forming impermeable nanocapsules when the ends of the CNTs are closed.29, 30 Having closed ends allows the removal 4 of non-encapsulated material, after the filling step, under harsh washing conditions31 and even the retention of gases in their interior.32 Mutual protection is therefore afforded to both biological systems and the nanotube cargo delivered.7 When radionuclides are sealed, the action of ionizing particles exerts through the walls of CNTs25, 27, 28 and it spans from micron to millimeter distances depending on the characteristics of the chosen isotopes. It is worth noting that the internal radionuclides and external moieties (biocompatible/targeting molecules) can be independently changed thus giving versatility to this approach. In current clinical cancer therapy, half of all cancer patients are treated with radiation therapy either alone or combined with other types of therapies. Nanotechnology strategies however have been mainly driven by the use of chemotherapeutic agents. Notwithstanding, there is a growing interest in the development of nanomaterials bearing radioisotopes, since they offer a unique platform not only for molecular imaging but also to enhance the radiation response of tumors while reducing side effects.33-36 CNTs have been shown to deliver positron/γ-radioemitters and alpha particles of interest for imaging and therapeutic applications. This has been mainly achieved via chelation of radionuclides onto the external walls37-40 although the impregnation of radionuclides onto highly defective CNTs, which can be regarded as mesoporous materials, has also been investigated by Wilson et al.41-43 As just mentioned, an attractive feature of CNTs is the possibility to hermetically seal radionuclides in their interior. The nanocapsule will not only prevent release of the chosen radionuclides into the biological milieu but also of their decay products, which might be of a different chemical nature and detach from the nanocarrier with the previously employed strategies. We have previously reported a pioneering work using hermetically sealed Na125I-filled and externally glycosylated single-walled CNTs (SWNT).25 Specific tissue accumulation (lung) coupled with high in vivo stability prevented leakage of 5 radionuclide to high-affinity organs (thyroid/stomach) or excretion, and resulted in ultrasensitive imaging and delivery of an unprecedented radiation dose density25 Pascu et al. have recently given versatility to this approach and investigated the behavior of radiometal-filled (64Cu) SWNT in vitro and in vivo.27 In this study the ends of the CNTs were sealed using fullerenes as corks to prevent release of the encapsulated radioemitter.44, 45 In contrast to these previous reports, where ‘hot’ radionuclides, in the form of Na125I or 64 Cu(OAc)2, were directly filled into SWNT, followed by end-closing, herein, we present a novel approach by which ‘cold’ non-radioactive 152 Sm is initially sealed into the cavities of both SWNT and multi-walled CNTs (MWNT) leading to the formation of closed-ended 152Sm-filled carbon nanocapsules. The encapsulated and 152 153 stable Sm can then be activated into therapeutically active ‘hot’ Sm by neutron irradiation through the walls of the CNTs (Figure 1a; Schematic representations comparing both strategies to achieve ‘hot’ nanocapsules are included in Figure S1). After neutron activation, exceptionally high specific activities were obtained, compared to previous works on direct encapsulation of radionuclides, which allowed not only in vivo nuclear imaging but also lung tumor radiotherapy. 153 Sm is an attractive radionuclide as it emits beta particles at maximum energy of 810 keV (suitable for cell killing), and also releases gamma energy of 103 keV (allowing clinical imaging).46, 47 153 Sm is used clinically in the form of 153 Sm-ethylene diamine tetramethylene phosphonate chelate (153Sm-EDTMP, Quadramet®) for palliation of bone metastases.48 It offers added advantages such as desirable half-life (46.3 h) and lower gamma energy over other therapeutic radionuclides currently being used in clinic such as iodine-131 or yttrium-90.49 The encapsulation of ‘hot’ radionuclides into the cavity of CNTs requires a fast and safe manipulation of the material due to their constant decay. By using the neutron activation strategy (Figure 1a), both the filling and the removal of the non-encapsulated compound do not require 6 the use of radioactive facilities thus reducing radiation exposure to the personnel. It also alleviates the time constraints imposed by the constant decay of radionuclides since the ‘cold’ nanocapsules can be shelf-stored and activated by neutron irradiation on demand. Neutron activation of nanoparticles incorporating neutron-activatable isotopes is well documented, with initial studies focused on the use of fullerenes and polymeric nanoparticles.50-54 Di Pasqua et al. also proposed the use of inorganic mesoporous silica nanoparticles loaded with neutron- activatable holmium-165.55, 56 Taking advantage of the porosity of the nanocarrier, this was later expanded to chemoradiotherapy.57 Other inorganic materials including holmium iron garnet, mesoporous carbon nanoparticles and graphene oxide have more recently been explored as holmium nanocarriers.58-60 On the other hand, neutron activation of large particles (i.e. holmium- 166 containing microspheres) has been employed in clinical trials for the treatment of liver tumor via intra-arterial injection leading to radioembolization.61, 62 In this work we propose the use of 152 neutron irradiation of ‘cold’ isotopically enriched Sm sealed within carbon nanotubes. The activated ‘hot’ 153Sm nanocapsules allow both in vivo imaging and tumor radiotherapy. RESULTS AND DISCUSSION 152 Preparation and neutron activation of Sm-filled nanocapsules to yield radioactive 153 Sm@CNTs. As-received CNTs were initially purified and shortened by combination of solution processing with strong acids followed by steam.63 High temperature molten phase 152 capillary filling was employed for the encapsulation of isotopically enriched SmCl3 into SWNT and MWNT, leading to the closing of the nanotube ends on cooling29, 30 (Figure 1a). For 152 ease of nomenclature, we will refer to these nanocapsules as Sm@SWNT and 152 Sm@MWNT. We employed enriched samarium (152Sm) to ensure an efficient activation to the 7 153 clinically employed Sm isotope upon neutron irradiation.64 After the filling experiment, the 152 excess of SmCl3 external to the walls of the CNTs was removed by repeated washings and filtration of the sample in water. The length and diameter distribution of the samples was determined after the filling step. The median length turned out be 199 nm for SWNT and 331 nm for MWNTs (see Figure S2 and Table S1 for statistical data), and the median diameters 2.1 nm for SWNTs and 10.7 nm for MWNTs (Figure S3, Table S2). The amount of samarium present was next quantitatively determined by ICP-MS. The analyses revealed ~12.3 w/w% and 17.6 w/w% loading of 152Sm for 152Sm@SWNT and 152Sm@MWNT, respectively. The bigger cavity 152 of Sm in MWNTs with respect to their single-walled counterparts accounts for the larger 152 152 amount of encapsulated Sm. Due to the large excess of SmCl3 (10 mg) that was employed for the bulk filling of CNTs (1 mg), only a small fraction of the metal halide was encapsulated (ca. 2 % according to the filling yield). With the present strategy, the non-encapsulated 152SmCl3, dissolved during the washing steps, is recovered by filtration. This compound is an enriched non- radioactive form of samarium which can be dried (from the collected filtrate), stored and employed in further filling experiments. Therefore, no samarium waste is generated in the filling step. Neutron activation is favored in radionuclide production because of its relatively simple process and the availability of various types of nuclear reactors. It also permits lower radiation exposure to the personnel since the preparation procedures such as encapsulation, purification and characterization can be conducted prior to irradiation. Exceptionally high specific radioactivities (SRA) with respect to the mass of the composite material were obtained after 96 h 152 152 neutron irradiation of Sm@SWNTs (6.33 GBq/mg) and Sm@MWNT (11.37 GBq/mg) (Table 1). The SRA values are two orders of magnitude (differ by a factor of about 100) higher 8 than those previously reported for neutron irradiated nanocarriers loaded with clinically relevant radiotherapeutic isotopes, which are in the range of ca. 1-80 MBq/mg (Table S3).65 This is due to the high neutron flux and long irradiation time employed in the present study taking into 152 account the amount of Sm loading. In fact, such high SRA are in line of those achieved by neutron irradiation of pure elements or their oxides for industrial applications.64 Structural examination of the nanocapsules before (152Sm@CNT) and after (153Sm@CNT) neutron irradiation. Instant thin-layer chromatography (ITLC) examination was 153 153 performed for Sm@SWNT and Sm@MWNT to evaluate whether neutron irradiation had 152 caused any structural damage to the carbon nanocapsules that could result in leakage of Sm from their interior. Based on recent studies on non-enriched SmCl3 nanocapsules, over 99% of SmCl3 that is not sealed inside the CNTs can be easily dissolved by a short sonication (10 s) of the sample in water.66 Therefore, to ensure the complete dissolution of 153 Sm from the inner cavities of those CNTs that could have undergone structural damage, a longer sonication step (several min) in water was performed. As shown in Table 1 and Figure S4 a remarkable amount 153 of Sm remained sealed in the cavities of MWNT (92%) and SWNT (70%), despite the high neutron flux (1.6x1014 n cm-2s-1) and long neutron irradiation time employed (96 h) to achieve high SRA. Milder irradiation conditions are usually used to prevent/decrease degradation of the carrier during neutron activation. Comparison of SRA under different conditions is included in 153 Table S1. The fraction of Sm external to the CNT walls could be either chelated prior to in vivo studies or washed from the sample resulting in a minimal amount of radioactive waste, compared to the previously employed strategy where radionuclides were directly filled into CNTs. The presence of several concentric graphene layers resulted in a more stable construct to the flux of neutrons. Therefore, although 152Sm can be activated inside both MWNT and SWNT, 9 the former turned out to be more robust for the preparation of ‘hot’ nanocapsules by neutron 152 irradiation. Even higher fractions of permanently sealed Sm would be expected in both constructs by decreasing the time and flux of neutron irradiation, with the corresponding reduction of SRA (according to equation (1)). Transmission electron microscopy (TEM) allowed us to evaluate the effect of neutron 152 152 irradiation on the Sm-filled CNTs. Encapsulated SmCl3 can be observed in the high 152 152 resolution TEM (HRTEM) images of Sm@SWNT and Sm@MWNT before irradiation (Figure 1b, top panels). The filled metal halides were still present in both types of CNTs (black arrows) after irradiation without evident damage on the nanotube structure. It is important to note that the ends of the carbon nanotubes remain sealed after neutron irradiation (pointed with white arrows). Additional TEM images for both ‘cold’ and ‘hot’ nanocapsules are included in Figure S5. Structural changes of the encapsulated crystals are sometimes observed during electron microscopy imaging.67-72 In the present case we observed that in some cases, SmCl3 nanowires were fragmented into smaller crystals (Figure S6). Preservation of the payload encapsulation was further confirmed by high angle annular dark field (HAADF) scanning transmission electron microscopy (STEM) (Figure 1b, central panels). The SmCl3 filling (bright white lines) aligned nicely within the CNT lumen (dim grey contrast) even following irradiation also indicating the presence of filling material. Energy dispersive X-ray (EDX) spectroscopy was conducted to assess the preservation of the payload content upon irradiation. The presence of samarium was detected before and after 153 irradiation (Figure 1b, bottom panels). Europium, the decay product of Sm, was not detected by EDX because despite the high SRA less than 1% of 152Sm nuclei underwent nuclear reaction to radioactive 153Sm under the employed conditions. 10 153 Whole body SPECT/CT imaging and tissue biodistribution of Sm@CNTs in normal 153 mice following i.v. injection. The Sm@CNT suspensions underwent different dilutions with non-irradiated 152Sm@CNT to contain the appropriate radioactivity for the in vivo studies in 1% Pluronic® F-127 saline. As just discussed, stable nanocapsules sealing 70 % (SWNT) and 92 % (MWNT) of 153Sm were obtained. The sonication step after neutron irradiation ensures no further 153 153 leakage of Sm from the nanocapsules in vivo. The fraction of Sm dissolved during the preparation of ‘hot’ nanocapsules was chelated with EDTA to ensure a fast renal clearance.73 The pharmacokinetics and biodistribution of 153Sm@SWNT and 153Sm@MWNT up to 24 h were next explored by whole body SPECT/CT imaging and quantitative γ-scintigraphy (Figure 2). 153 SPECT/CT imaging demonstrated that both Sm@CNTs showed similar biodistribution patterns with accumulation mostly in spleen, lung and liver within 30 min (Figure 2a). The uptake in these organs was retained up to 24 h post injection (period of imaging). We have previously observed in vivo lung accumulation of filled CNTs bearing targeting and non- targeting groups.25, 28 This behavior has been attributed to the physical properties of CNTs within physiology.28 As expected, high signals were detected from kidney and bladder within 30 min 153 after injection, which were attributed to the Sm:EDTA chelate. From quantitative γ- scintigraphy measurements, a progressive clearance from the blood was observed for both 153 Sm@CNTs, in which most of the compounds were cleared out from circulation at 4 h post injection (Figure 2b). Figure 2c shows roughly 20% and 10% of injected dose (ID) were detected 153 153 in urine in the case of Sm@SWNT and Sm@MWNT, respectively. The results correlated 153 well with the ITLC findings where a higher amount of free Sm was detected after neutron 153 153 irradiation in Sm@SWNT than in Sm@MWNT (Figure S4). Negligible radioactivity was 153 detected in feces for both Sm@CNTs (< 0.5 %ID). The organ biodistribution profiles 11 established at 4 and 24 h (Figure 2d) paralleled the SPECT/CT imaging results where 153 Sm@CNTs largely accumulated in spleen and showed high uptake in lung and liver. The accumulation in spleen and liver increased over time while the uptake in lung at 4 h and 24 h remained steady. Biodistribution analyses expressed as %ID per organ are presented in Figure S7. 153 Assessment of radiotherapeutic efficacy of Sm@CNTs in B16F10-Luc melanoma lung metastasis. Taking advantage of the prominent lung accumulation property of CNTs, the 153 153 radiotherapeutic efficacy of Sm@SWNT and Sm@MWNT was investigated using an experimental B16F10-Luc melanoma lung metastatic tumor model. This syngeneic tumor model is characterized as highly aggressive and metastatic, and has been commonly used for evaluation of cancer regimens.74, 75 The luciferase-expressing feature of this cell line enabled tumor growth monitoring by whole body bioluminescence imaging. The results demonstrate that single intravenous administration of 20 MBq of 153Sm@SWNT or 153 Sm@MWNT (200 µg of CNTs per mouse) significantly delayed the tumor growth in lung 153 (Figure 3). Representative bioluminescence images of untreated, Sm@SWNT and 153 Sm@MWNT treated mice on day 10 and day 16 post-tumor inoculation are shown in Figure 3a and images of all mice are shown in Figure S8. From the lung tissues examined at post- mortem (Figure 3b), distinct differences were visualized between the treatment groups. Metastatic melanoma nodules occupied most of the lung tissue from the untreated mice, while lesser number of melanoma nodules were observed in 153Sm@CNT treated mice. The number of tumor nodules also reflected the lung weights in which lung tissues from untreated mice weighed significantly heavier than 153Sm@CNT treated mice (***p<0.001). 12 Figure 3c displays the average tumor size (photons/sec) of untreated mice and mice treated 153 153 with Sm@SWNT or Sm@MWNT over the therapy time course. Treatment responses 153 started to appear on day 13 for both Sm@CNTs showing significant tumor reduction compared to the untreated group (*p < 0.05). Although similar amount of radioactivity was 153 administered and both Sm@CNTs showed comparable % ID in lung, notably smaller tumors 153 153 were measured in mice treated with Sm@SWNT compared to Sm@MWNT on day 16 (*p 153 < 0.05). Nevertheless, both Sm@CNT treatments demonstrated substantial tumor growth suppression on day 16 (***p < 0.001, compared to control) where the therapy study terminated. 153 The reasons behind this more prolonged therapeutic effect of Sm@SWNT are not clear, but 153 there might be differences between the two Sm@CNTs in terms of the interaction with lung tissues/tumor cells which led to the differential therapeutic outcomes. Histological analysis was carried out on H&E (hematoxylin and eosin stain) and Neutral Red stained lung tissue sections at post-mortem at the end of therapy study on day 16. The lung sections from untreated mice showed an invasion of cancerous mass leading to the disappearance of healthy lung tissues and a clear reduction of alveoli number (Figure 3d, top panels). In 153 contrast, in Sm@CNT treated mice (Figure 3d, middle and bottom panels), lung sections appeared healthier with the presence of fewer colonies of melanoma nodules. Zoom-in images of healthy and cancerous areas from each treatment groups are shown on the right panels with 153 arrows indicating some trapped Sm@CNTs. CNTs were difficult to identify in cancerous regions due to the dark appearance of melanoma. Together with the bioluminescence images, it 153 indicated that Sm@CNT therapies did not fully eradicate tumors but prevented their proliferation. 13 Toxicity assessments post 153 Sm@CNT injection and after radiotherapy. During the radiotherapy period, variable whole body weight changes were observed within groups (Figure 153 4a). Body weight loss seemed to occur for the Sm@CNT treated mice starting from day 10 post-tumor inoculations but there is no significant difference between the groups including the untreated mice. The weights of major organs were measured at sacrifice. Marginal changes were 153 found between most of the untreated and Sm@CNT treated organs, except for lungs (with tumors, ***p < 0.001) and spleen (***p < 0.01 or ***p < 0.001) (Figure 4b). As previously evidenced in Figure 3b, the aggressively growing tumors increased the weight of lung significantly for the untreated mice. We have previously demonstrated that cold metal-filled CNTs were biocompatible in vivo.76 The significant weight loss in lungs from mice treated with 153 Sm@CNTs compared to untreated mice is likely a result of reduction of tumor nodules. Histological assessment performed on mice treated with low dose radioactivity (200 µg 153 Sm@CNTs/mouse, 1 MBq) showed no significant histological abnormalities in the selected major organs (heart, lung, kidneys, liver, and spleen) sampled at 24 h post injection (Figure S9). This further confirms that the reduced lung weights were due to the radiotherapy effect and not from the CNTs per se. The average weight of spleen of 153Sm@CNT treated mice was reduced compared to untreated tumor-bearing mice. This is less likely to be general toxicity of the CNTs but a consequence of the high spleen uptake of 153Sm@CNTs leading to high radioactivity accumulation. Histological examination on spleen with profound observation of the death of splenocytes could explain the 153 spleen and body weight reduction of Sm@CNT treated mice since the malfunction of spleen can debilitate the immune system (Figure 4c). No histological abnormalities were observed for heart, liver, and kidney. The lack of liver toxicity further supports containments of 153Sm within 14 the CNTs in vivo. For example, studies by other groups demonstrated hepatotoxicity with 153 unchelated Sm which naturally ends up in reticuloendothelial organs such as the liver and spleen.48, 73 Hematological and inflammatory examinations are needed in future investigations. 153 Dosimetry simulation studies. Since the developed therapeutic Sm@CNT nanocapsules achieved promising killing effect on tumor cells, a model was established to simulate the performed radiation therapy. Figure 5 shows the comparison between the experimental (green dots) and simulated tumor growth (blue lines, according to equation (2)) data of untreated mice, 153 153 153 and mice treated with Sm@SWNT or Sm@MWNT. The Sm@CNTs were injected on day 8 which is the reason for the hump on the simulated curve at day 8. For both 153Sm@SWNT 153 and Sm@MWNT the effect of irradiation was to decrease the rate of tumor growth. Despite the many simplifying assumptions of the model (e.g. mono-exponential uptake, washout, and proliferation rates), the simulations reproduced the general trend of the experimental data for both systems. Moreover, the simulations successfully predicted that, for the particular administered activity, biodistribution features and tumor cell type, the resulting dose rate was not high enough. The performed radiotherapy clearly decreased the rate of tumor growth; however, the cell killing could not dominate over cell proliferation which could lead to complete tumor regression. The simulations suggest that local tumor control (defined here as the eradication of all tumor cells with 95% probability) requires roughly 17-28 times higher administered activities 153 153 (330 MBq for Sm@MWNT and 550 MBq for Sm@SWNT). Taking into account the obtained SRA values for the prepared nanocapsules, it is indeed possible to administer the required dose for complete eradication of cancer in vivo (ca. 30–90 µg of undiluted 153 Sm@CNT). In view of efficacy and safety, toxicity profiles using higher doses should be further studied. Due to radiation regulation limits, a much lower dose (20 MBq) has been 15 employed herein for the internal radiation therapy studies. Remarkably, even using such small doses significant delay in tumor growth has been observed. CONCLUSIONS The development of radiopharmaceuticals requires a simple and safe handling of raw materials and final products, during both the manufacturing process, at the point of care, and during the intermediate steps. Furthermore, radiopharmaceuticals must be chosen carefully to deliver the correct activity for the particular radionuclide, given its’ half-life. The current work is an original 153 concept to produce highly energetic Sm-encapsulated CNTs via a stable intermediate 152 consisting of Sm-filled CNT nanocapsules, that may ultimately be used in a therapeutic context. Preparing the CNT nanocapsules in this way enables control over the 153Sm content and consequently, the activity. Furthermore, it reduces the process-complexity and the amount of waste that would be generated if one performed the encapsulation process directly on radioactive isotopes. Finally, the isotopically enriched ‘cold’ nanocapsules can be stored and activated by neutron irradiation when needed. Neutron irradiation of ‘cold’ 152Sm-nanocapsules turned out to 153 be robust, especially for MWNT with 92% retention of Sm. Extremely high SRA were 153 153 obtained (6.33 GBq/mg for Sm@MWNTs and 11.3 GBq/mg for Sm@SWNTs), which allowed live imaging and therapy in experimental metastatic lung tumor model in mice. The obtained SRA are two orders of magnitude higher than those previously reported for neutron activated nanomaterials bearing clinically relevant radioisotopes.65 On the other hand, the activity of a standard dose of clinically employed yttrium-90 microspheres is 2-4 GBq.77-79, requiring the administration of over 100 mg (52.0 mCi in 116 mg TheraSphere®).80 Remarkably, this activity can be achieved with less than 1 mg of the developed nanocapsules. Dosimetry 16 modelling studies can further optimize the treatment efficacy for potential clinical applications and predict the dose required. The preparation of nanocapsules with enriched isotopes benefits from well established filling and end closing strategies, thus giving versatility in the selection of neutron activatable imaging and/or therapeutic isotopes. The possibility of carrying out further surface functionalization on the irradiated nanocapsules can potentially result in a stable nanoconstruct capable of in vivo targeting, imaging, and therapy. METHODS Additional experimental details are included in the Supporting Information. Preparation of ‘cold’ nanocapsules (152Sm@CNT). Both SWNTs and MWNTs (Elicarb®) were initially treated to shorten the tubes, open their ends and remove carbonaceous and metallic (catalyst) impurities. SWNTs were exposed to a combined piranha-steam treatment, whereas MWNTs underwent a combined H2SO4:HNO3-steam treatment following previously reported protocols.63 Steam is very efficient in removing functional groups and highly defective carbon nanotubes.81 Enriched 152 Sm2O3 (CIS-Bio International-Ion Beam Applications, France) was 152 transformed to anhydrous SmCl3 following the protocol reported for the synthesis of anhydrous SmCl3 with natural isotopic distribution (non-enriched).82 The synthesized anhydrous 152 SmCl3 is highly hygroscopic and was handled under an inert atmosphere, and employed for 152 filling CNTs. SWNT or MWNT and SmCl3 were ground together in a weight ratio 1:10 (CNTs:152SmCl3) inside an argon filled glove box, split in smaller fractions, placed inside silica tubes and sealed under vacuum. The mixtures were annealed for 12 h at 900 °C (SWNT) or 1200 °C (MWNT) thus leading to the formation of carbon nanocapsules (closed-ended filled CNTs).29, 30 Removal of the non-encapsulated 152 SmCl3 with 0.6 M HCl was followed by UV- 17 152 Vis, until no more SmCl3 was detected in the washings (0.2 μm Whatman® polycarbonate membranes).66 The length distribution of the resulting 152 Sm@SWNT and 152 Sm@MWNT was determined from SEM images, following a previously described methodology,83 and the diameter distribution from TEM images. Two hundred CNTs were measured for each of the analyses (see SI). Neutron activation of 152Sm-filled CNTs to achieve ‘hot’ nanocapsules (153Sm@CNT). The irradiation protocol was established according to equation (1):64 . σΦ λ A 1 (1) where A is the predicted activity of the radioisotope produced (Bq g-1), M is the atomic mass of the target element (152 g mol-1 for 152Sm), Φ is the neutron flux of the reactor (1.6x1014 n cm- 2 -1 s ), σ is the thermal neutron activation cross-section of the target isotope (206 barns for 153Sm), λ is the decay constant (0.693/T1/2), (T1/2 is the half life of the target isotope, which is 46.27 h for 153 Sm) and t is the irradiation time (96 h). Vacuum sealed silica ampoules containing 30 mg of either 152Sm@SWNT or 152Sm@MWNT were placed in an aluminum capsule and irradiated at a neutron flux of 1.6x1014 n·cm-2·s-1 for 96 h in a pool-type reactor (OSIRIS, CEA Saclay, France). 153Sm@SWNT and 153Sm@MWNT powders were separately suspended in a volume of 1% Pluronic® F-127 saline (0.9 % NaCl) solution at 0.5 mg/mL. Characterization. HAADF-STEM images were acquired at 20 kV on an FEI Magellan XHR Scanning Electron Microscope (SEM). HRTEM images were acquired on a FEI Tecnai G2 F20 microscope at 200 kV. Samples were dispersed in ethanol and deposited onto lacey carbon Cu grids. 153Sm@CNT were imaged after the complete decay of radioactivity. EDX was carried out on a FEI Quanta SEM microscope equipped with an EDAX detector at 20 kV. EDX spectra were acquired on large areas of CNTs (> 50 m2) to obtain representative data of the composition of 18 the samples. ICP-MS analysis was carried out using an ICP-MS with a quadrupole collision cell (PerkinElmer Sci EX ELAN® DRC II). Samples for ICP-MS were microwave digested (800 W, 40 min) in a 3:1 mixture of nitric:hydrogen peroxide (Suprapur® Merck KGaA, Germany). For 153 153 ITLC, aliquots of Sm@SWNT and Sm@MWNT were spotted on TLC strips (Agilent Technologies, UK) and then developed in 0.1 M ammonium acetate containing 50 mM EDTA as a mobile phase. Strips were allowed to dry and counted quantitatively using a cyclone phosphor detector (Packard Biosciences, UK). In vivo studies. All in vivo experiments were conducted under the authority of project and personal licenses granted by the UK Home Office and the UKCCCR Guidelines (1998). Female C57BL/6 mice aged 6-8 weeks were purchased from Harlan (UK) and used for all in vivo studies. Mice were anaesthetized by isoflurane (IsoFlo®, Abbott Laboratorie Ltd, UK) inhalation and injected via a tail vein. Each injection dose contained the same amount of CNT (200 µg), and the necessary radioactivity (for imaging, biodistribution or therapy studies) was a by dilution of the 153Sm@CNT suspensions with non-irradiated 152Sm@CNT (2 mg/mL in 1% Pluronic® F- 127 saline). Injection suspensions were added with 0.1 M EDTA (one twentieth of the injection 153 volume) to chelate any free Sm. For histological examination, collected organs were fixed in 10% neutral buffered formalin. Harvested fixed organs were paraffin-embedded and sectioned for haematoxylin and eosin (H&E) or Neutral Red staining according to standard histological protocols at the Royal Veterinary College (UK). All stained sections were analysed using a Leica DM 1000 LED Microscope (Leica Microsystems, UK) coupled with a CCD digital camera (Qimaging, UK). 153 In vivo imaging, pharmacokinetics and organ biodistribution of Sm@CNTs. Biodistribution was firstly assessed by 3D whole body SPECT/CT imaging (Nano-SPECT/CT, 19 Bioscan, USA). SPECT images were taken in 24 projections over 30-40 min using a four-head scanner with 1.4 mm pinhole collimators. CT scans were performed at the end of each SPECT acquisition. Images were reconstructed by MEDISO software (Medical Imaging Systems), and SPECT and CT images were merged using the InVivoScopeTM software (Bioscan, USA). The radioactivity of tissues, blood, urine or feces was measured by γ-scintigraphy (LKB Wallac 1282 Compugamma, PerkinElmer). Blood (sampled from a tail vein), urine and feces were collected over 24 h. To assess the excretion profiles, animals had free access to water but not food. 153 In vivo Sm@CNT radiotherapy studies on B16F10-Luc lung metastasis tumor model. Mice were injected with 5 x 105 B16F10-Luc cells in 0.2 mL of PBS via a tail vein to establish pulmonary melanoma metastases. Following tumor inoculation, in vivo quantitative bioluminescence imaging was performed on day 7, 10, 13 and 16 to monitor the tumor growth (IVIS Lumina III, Perkin-Elmer, UK). Mice were subcutaneously injected with luciferin (D- luciferin potassium salt, Perkin-Elmer, UK) at 150 mg/kg and imaged 10 min after injection. To 153 determine the radiotherapeutic action of Sm@CNT, B16F10-Luc tumor bearing C57BL/6 153 mice were randomly divided into three groups (n = 9-10): untreated, Sm@SWNT and 153 Sm@MWNT. On day 8 post tumor inoculation, mice were injected with 153Sm@CNT (ca. 20 MBq). Mice were sacrificed on day 16 post tumor inoculation, and major organs were collected. Tumor growth was monitored by bioluminescence imaging. The tumor growth data was expressed as mean ± SEM (standard error of the mean), with n denoting the number of animals. Dosimetry simulation. For targeted radionuclide therapy, the surviving fraction (SF) of the tumor cell population over time, SF ( t ) , can be described by equation (2).84   m  k 1 m t 1 k t   SF(t )  exp  r0   e  e    t  (2)   mk m k   20 where  is the radiosensitivity parameter (Gy-1), r0 is the extrapolated initial dose rate (Gy h-1), m is the radionuclide effective uptake rate (h-1), k is the radionuclide effective washout rate (h- 1 ) and  is the tumor cell proliferation rate (h-1). The effective rates of uptake and washout of the radionuclide account for both biological and physical processes and they are defined through m ln( 2 ) Tu  ln( 2 ) Tp and k  ln( 2 ) Tw  ln( 2 ) Tp , respectively, where Tu is the biological uptake half-life, Tw is the biological washout half-life, and T p is the physical half-life of the radionuclide. The values of m and k were obtained from a fit of the experimental tumor biodistribution data for the percentage injected dose (% ID) using equation (3):  ID(t )  ID0 e k t  e  m t  (3) where ID0 is the extrapolated % ID in the tumor at time t  0 . Note that the physical half-life of 153 Sm ( Tp  46.3 h ) sets a lower limit to the values of m and k which should be larger than ln(2 ) Tp . The least square fits yielded ID 0  16 .6 % , k  0.0162 h 1 , m  2.32 h 1 for SWNTs, and ID 0  28 .0 % , k  0.0160 h 1 , m  0.826 h 1 for MWNTs (Figure S10). To obtain the rate constant for tumor cell proliferation (  ), we assume (following standard practice) that all tumor cells in the population are proliferating. The experimental tumor growth data of the non-irradiated group of mice are fitted in equation (4): N (t )  N 0 e t , (4) where N 0 is the number of tumor cells at t  0 . The rate constant  in equation (4) is normally expressed as   ln(2 ) Td with Td being the tumor doubling time. Fit of the experimental data by equation (4) yields Td  39.78 h (see Figure 7a). 21 The extrapolated initial dose rate in the tumor, r0 , entering equation (2) is estimated from equation (5): r0  A0 ID 0 S f T , (5) where A0 is the administered activity to the mice (20 MBq), ID0 is obtained from equation (3), fT is the fraction of the organ activity that is taken up by the tumor, and S is the radiation absorbed dose to the tumor per radionuclide decay (Gy Bq-1 s-1). For 153Sm and a tumor mass of 0.07 g, S  4.25 10 10 Gy Bq 1 s 1 .85 To obtain fT , it is assumed that the activity was distributed uniformly in the tumor ( M tumor  0.07 g ) and the lungs ( M lungs  0.12 g ), so fT   0.368 . Substituting the above parameters in equation (2), the values M tumor M lungs  M tumor r0  3.158 Gy h 1 and r0  1.872 Gy h 1 are obtained for MWNTs and SWNTs, respectively. The radiosensitivity parameter α was determined empirically from cell survival data using the expression ln( SF2 )  , (6) D  D 2 ( a /  ) 1 where SF2 is the measured surviving fraction at D  2 Gy and  /  is the dose where the linear and quadratic cell-kill mechanisms (of the linear-quadratic cell survival model) are equal. For B16F10-Luc cells, SF2  0.96 .86 Such a high value for SF2 corresponds to an extremely radioresistant cell line. Thus, to a first approximation, we can assume  /   1 Gy which is at the low-end of the range of values observed for the late responding tissues.87, 88 equation (6) yields   0.0068 Gy-1. From the above dosimetry model we may also obtain a rough estimate of the required administered activity for local tumor control, defined here as the eradication of all (clonogenic) 22 tumor cells with 95% probability. Assuming that 70 mg of tumor contain roughly 7×107 cells (and all are clonogenic) the above local tumor control probability translates to the condition . 153 7 10 which can be achieved with 670 MBq Sm@SWNT and 400 MBq 153 Sm@MWNT administered activity. If we consider the (perhaps) more realistic approximation . that only 1% of tumor cells are clonogenic (i.e., 7 10 ) the corresponding administered activity values are 550 MBq 153Sm@SWNT and 330 153Sm@MWNT. ACKNOWLEDGMENTS This work received funding from the European Union’s Seventh Framework Programme (FP7- ITN Marie-Curie Actions, RADDEL, 290023). K.T. Al-Jamal acknowledges funding from Worldwide Cancer Research (12-1054), M. Kalbac from MEYS (LL1301 and LTC18039) and G. Tobias from Agaur (2017 SGR 581). ICMAB and ICN2 acknowledge financial support from the Spanish Ministry of Economy and Competitiveness, through the “Severo Ochoa” Programme for Centres of Excellence in R&D (SEV-2015-0496 and SEV-2017-0706). This work was partly supported by the Centre National de la Recherche Scientifique (CNRS), by the Agence Nationale de la Recherche (ANR) through the LabEx project Chemistry of Complex Systems (ANR-10- LABX-0026_CSC) and by the International Center for Frontier Research in Chemistry (icFRC). We thank Thomas Swan Co. Ltd. for supplying CNT Elicarb® samples. We are grateful to S. Sandoval (ICMAB) and C. Ramos (IQS-ICMAB) for assessing the length and diameter distribution of CNTs. 23 FIGURES Figure 1. Preparation and characterization of carbon nanocapsules. a) Schematic representation of the strategy employed for the preparation of ‘hot’ nanocapsules. A non-radioactive enriched precursor is initially encapsulated (152Sm) and it is activated into its radioactive form by neutron irradiation (153Sm) in the last step. The wavy lines indicate radioactivity emerging from radionuclides (153Sm). b) Structural characterization of Sm-filled CNTs before (‘Cold’ nanocapsules) and after (‘Hot’ nanocapsules) irradiation. HRTEM images are shown in the top 24 panels. Black arrows point to the SmCl3 filling, and white arrows to closed tips of carbon nanotubes. The crystalline structure of SmCl3 can be better appreciated on the MWNT images, with visible lattice fringes (parallel lines). HAADF-STEM images are shown in the middle panels. The bright lines correspond to the filling of SmCl3, while CNTs appear as a dim grey contrast. EDX analyses confirming the presence of SmCl3 are shown in the bottom panels. 25 153 Figure 2. Whole body SPECT/CT imaging and tissue biodistribution of Sm@SWNT and 153 Sm@MWNT in mice up to 24 h. a) SPECT/CT imaging, b) blood circulation, c) excretion profiles and d) tissue biodistribution profiles. C57BL/6 mice were i.v. injected with 200 µg of 153 153 EDTA-quenched Sm@SWNT or Sm@MWNT containing 10 MBq or ~1 MBq for SPECT/CT imaging or γ-scintigraphy, respectively. In excretion studies, mice were housed individually in metabolic cages for 24 h immediately after injection. The radioactivity in blood, major organs sampled at specified time points and urine/feces collected at 24 h post injection, was measured by γ-scintigraphy. The results are expressed as mean ± S.D. (n=3-4). 26 Figure 3. Tumor growth delay studies in experimental metastatic lung tumor mice model after internal administration of radiotherapy. B16F10-Luc tumor-bearing C57BL/6 mice received a single i.v. injection of 153Sm@SWNT or 153Sm@MWNT (20 MBq, 200 µg) on day 8 post-tumor inoculation. a) Average lung weights measured on day 16 post-tumor inoculation, the 27 153 experimental endpoint. b) Representative whole-body images of untreated, Sm@SWNT and 153 Sm@MWNT treated mice captured on days 10 and 16 post-tumor inoculation. c) Tumor growth monitoring over time. Bioluminescence signals correspond to luciferase expressing B16F10 cells in the lung. d) H&E and Neutral Red stained B16F10-Luc tumor-containing lung tissue sections excised on day 16 (left panels). High magnification images of tumor (areas 1) and non-tumor (areas 2) lung tissues are shown on the right. Arrows indicate the presence of CNTs. Scale bars: 200 μm for H&E and Neutral Red stained sections; 5 μm for zoom-in images of areas 1 and 2. Results are presented as mean ± S.D. (n=9-10). Significant differences were examined using one-way ANOVA followed by Tukey’s multiple comparison test (*p < 0.05, **p < 0.01, ***p < 0.001). 28 153 Figure 4. Toxicity assessments in tumor-bearing mice treated with Sm@SWNT and 153 Sm@MWNT via intravenous injection. B16F10-Luc tumor-bearing C57BL/6 mice received a single i.v. injection of 153Sm@SWNT or 153Sm@MWNT (20 MBq, 200 µg) on day 8 post-tumor inoculation. a) Percentage of whole body weight changes of B16F10-Luc tumor-bearing mice 29 after radiotherapy. b) Major organ weight and c) histological examination of major organs on day 16 post tumor inoculation. The body weight of individual mouse was measured every 3-4 days post tumor-inoculation. At the experimental endpoint on day 16, major organs including heart, kidney, liver and spleen were excised, weighed and fixed for histological examination. Tissue sections were stained with H&E (examine necrosis) or Neutral Red (track CNTs). Scale bars: 50 μm. Results are presented as mean ± S.D. (n = 9-10). No statistical differences were found in % 153 whole body weight changes between different treatments. Mice treated with Sm@CNTs showed significant weight reduction in lung and spleen. Significant differences were examined using one-way ANOVA followed by Tukey’s multiple comparison test (**p < 0.01, ***p < 0.001). 30 Figure 5. Predicted and experimental relative tumor sizes as a function of time in control or 153 153 Sm@SWNT- and Sm@MWNT-treated mice. B16F10-Luc tumor-bearing C57BL/6 mice 153 153 received a single dose i.v. injection of Sm@SWNT or Sm@MWNT (20 MBq, 200 µg) on day 8 post tumor inoculation or remained untreated. Green dots represent experiment data and 153 blue lines are simulated results for the three groups: a) untreated control, b) Sm@SWNT 153 treated group, and c) Sm@MWNT treated group. The calculated blue lines are plotted according to equation (2). The values of the coefficients entered in equation (2) are as follows: 31 for all three cases, the radiosensitivity parameter was =0.0068 Gy-1 and the cell doubling time Td equals 39.78 h which corresponds to a proliferation rate = ln(2)/Td = 0.017 h-1. Extrapolated initial dose rate r0, effective uptake rate m, and effective washout rate k are specific for each case. Specifically, for the untreated sample (a), the extrapolated initial dose rate r0 equals 0 whereas for SWNT (b) the parameter values are: r0 = 1.872 Gy h-1, k = 0.0162 h-1, m = 2.34 h-1 and for MWNT (c) the parameter values are r0 = 3.158 Gy h-1, k = 0.0160 h-1, m = 0.826 h-1. The detailed derivation of the parameter values for each case is given in the Dosimetry simulation section in the Supporting Information. TABLES. 152 152 Table 1. Neutron irradiation of Sm@SWNT and Sm@MWNT to achieve radioactive 153 Sm@SWNT and 153Sm@MWNT. 152 153 Sm@CNT Sm@CNT Predicted Observed %152Sm Neutron ITLC % Starting material SRA SRA (ICP-MS) irradiation [h] Purity [GBq/mg] [GBq/mg] 152 Sm@SWNT 12.3 96 7.77 6.33 70 152 Sm@MWNT 17.6 96 11.11 11.37 92 32 ASSOCIATED CONTENT Supporting Information. Schematic representation employed for the development of ‘hot’ nanocapsules. Length distribution of 152Sm @CNT samples. Diameter distribution of 152Sm @CNT samples. ITLC analysis. HRTEM images of Sm-filled SWNTs before and after irradiation. Effect of the electron beam on the filling material. Table with neutron activated isotopes loaded onto nanocarriers. Tissue biodistribution of 153Sm@SWNT and 153Sm@MWNT. Live bioluminescence images. Live bioluminescence images of untreated mice or mice treated with 153Sm@CNTs. Histological examination of major organs. Fit of the experimental biodistribution data (% ID) in the lung. Additional experimental details. This material is available free of charge via the Internet at http://pubs.acs.org. AUTHOR INFORMATION Corresponding Author *E-mail: gerard.tobias@icmab.es (G. Tobias), khuloud.al-jamal@kcl.ac.uk (K.T. Al-Jamal), belen.ballesteros@icn2.cat (B. Ballesteros). Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes The authors declare no competing financial interest. 33 REFERENCES 1. Hartshorn, C. M.; Bradbury, M. S.; Lanza, G. M.; Nel, A. E.; Rao, J.; Wang, A. Z.; Wiesner, U. B.; Yang, L.; Grodzinski, P. Nanotechnology Strategies to Advance Outcomes in Clinical Cancer Care. ACS Nano 2018, 12, 24-43. 2. Jin, Q.; Deng, Y.; Chen, X.; Ji, J. Rational Design of Cancer Nanomedicine for Simultaneous Stealth Surface and Enhanced Cellular Uptake. ACS Nano 2019, 13, 954-977. 3. Rodríguez-Nogales, C.; González-Fernández, Y.; Aldaz, A.; Couvreur, P.; Blanco-Prieto, M. J. Nanomedicines for Pediatric Cancers. ACS Nano 2018, 12, 7482-7496. 4. Björnmalm, M.; Thurecht, K. J.; Michael, M.; Scott, A. M.; Caruso, F. Bridging Bio– Nano Science and Cancer Nanomedicine. ACS Nano 2017, 11, 9594-9613. 5. Liu, Z.; Jiang, W.; Nam, J.; Moon, J. J.; Kim, B. Y. S. Immunomodulating Nanomedicine for Cancer Therapy. Nano Lett. 2018, 18, 6655-6659. 6. Hong, G.; Diao, S.; Antaris, A. L.; Dai, H. Carbon Nanomaterials for Biological Imaging and Nanomedicinal Therapy. Chem. Rev. 2015, 115, 10816-10906. 7. Serpell, C. J.; Kostarelos, K.; Davis, B. G. Can Carbon Nanotubes Deliver on Their Promise in Biology? Harnessing Unique Properties for Unparalleled Applications. ACS Central Sci. 2016, 2, 190-200. 8. Hernández-Rivera, M.; Zaibaq, N. G.; Wilson, L. J. Toward Carbon Nanotube-Based Imaging Agents for the Clinic. Biomaterials 2016, 101, 229-240. 9. Loh, K. P.; Ho, D.; Chiu, G. N. C.; Leong, D. T.; Pastorin, G.; Chow, E. K.-H. Clinical Applications of Carbon Nanomaterials in Diagnostics and Therapy. Adv. Mater. 2018, 30, 1802368. 10. Calatayud, D. G.; Ge, H.; Kuganathan, N.; Mirabello, V.; Jacobs, R. M. J.; Rees, N. H.; Stoppiello, C. T.; Khlobystov, A. N.; Tyrrell, R. M.; Como, E. D., et al. Encapsulation of Cadmium Selenide Nanocrystals in Biocompatible Nanotubes: Dft Calculations, X-Ray Diffraction Investigations, and Confocal Fluorescence Imaging. ChemistryOpen 2018, 7, 144- 158. 11. Al-Jamal, K. T.; Nerl, H.; Muller, K. H.; Ali-Boucetta, H.; Li, S. P.; Haynes, P. D.; Jinschek, J. R.; Prato, M.; Bianco, A.; Kostarelos, K., et al. Cellular Uptake Mechanisms of Functionalised Multi-Walled Carbon Nanotubes by 3d Electron Tomography Imaging. Nanoscale 2011, 3, 2627-2635. 12. Kafa, H.; Wang, J. T.-W.; Rubio, N.; Venner, K.; Anderson, G.; Pach, E.; Ballesteros, B.; Preston, J. E.; Abbott, N. J.; Al-Jamal, K. T. The Interaction of Carbon Nanotubes with an In vitro Blood-Brain Barrier Model and Mouse Brain In vivo. Biomaterials 2015, 53, 437-452. 13. Pantarotto, D.; Singh, R.; McCarthy, D.; Erhardt, M.; Briand, J. P.; Prato, M.; Kostarelos, K.; Bianco, A. Functionalized Carbon Nanotubes for Plasmid DNA Gene Delivery. Angew. Chem. Int. Ed. 2004, 43, 5242-5246. 34 14. Pantarotto, D.; Singh, R.; McCarthy, D.; Erhardt, M.; Briand, J.-P.; Prato, M.; Kostarelos, K.; Bianco, A. Functionalized Carbon Nanotubes for Plasmid DNA Gene Delivery. Angew. Chem. 2004, 116, 5354-5358. 15. Battigelli, A.; Ménard-Moyon, C.; Da Ros, T.; Prato, M.; Bianco, A. Endowing Carbon Nanotubes with Biological and Biomedical Properties by Chemical Modifications. Advanced Drug Delivery Reviews 2013, 65, 1899-1920. 16. Dinesh, B.; Bianco, A.; Menard-Moyon, C. Designing Multimodal Carbon Nanotubes by Covalent Multi-Functionalization. Nanoscale 2016, 8, 18596-18611. 17. Marega, R.; Bonifazi, D. Filling Carbon Nanotubes for Nanobiotechnological Applications. New Journal of Chemistry 2014, 38, 22-27. 18. Martincic, M.; Tobias, G. Filled Carbon Nanotubes in Biomedical Imaging and Drug Delivery. Expert Opin. Drug Deliv. 2015, 12, 563-581. 19. Kotchey, G. P.; Hasan, S. A.; Kapralov, A. A.; Ha, S. H.; Kim, K.; Shvedova, A. A.; Kagan, V. E.; Star, A. A Natural Vanishing Act: The Enzyme-Catalyzed Degradation of Carbon Nanomaterials. Accounts of Chemical Research 2012, 45, 1770-1781. 20. Ueda, M.; Seo, S.; Nair, B. G.; Müller, S.; Takahashi, E.; Arai, T.; Iyoda, T.; Fujii, S.-i.; Tsuneda, S.; Ito, Y. End-Sealed High Aspect Ratio Hollow Nanotubes Encapsulating an Anticancer Drug: Torpedo-Shaped Peptidic Nanocapsules. ACS Nano 2019, 13, 305-312. 21. Chen, X.; Zhang, Q.; Li, J.; Yang, M.; Zhao, N.; Xu, F.-J. Rattle-Structured Rough Nanocapsules with in-Situ-Formed Gold Nanorod Cores for Complementary Gene/Chemo/Photothermal Therapy. ACS Nano 2018, 12, 5646-5656. 22. Ijäs, H.; Hakaste, I.; Shen, B.; Kostiainen, M. A.; Linko, V. Reconfigurable DNA Origami Nanocapsule for Ph-Controlled Encapsulation and Display of Cargo. ACS Nano 2019, 13, 5959-5967. 23. Wang, S.; Liu, X.; Chen, S.; Liu, Z.; Zhang, X.; Liang, X.-J.; Li, L. Regulation of Ca2+ Signaling for Drug-Resistant Breast Cancer Therapy with Mesoporous Silica Nanocapsule Encapsulated Doxorubicin/Sirna Cocktail. ACS Nano 2019, 13, 274-283. 24. González-Aramundiz, J. V.; Peleteiro, M.; González-Fernández, Á.; Alonso, M. J.; Csaba, N. S. Protamine Nanocapsules for the Development of Thermostable Adjuvanted Nanovaccines. Mol. Pharm. 2018, 15, 5653-5664. 25. Hong, S. Y.; Tobias, G.; Al-Jamal, K. T.; Ballesteros, B.; Ali-Boucetta, H.; Lozano- Perez, S.; Nellist, P. D.; Sim, R. B.; Finucane, C.; Mather, S. J., et al. Filled and Glycosylated Carbon Nanotubes for in Vivo Radioemitter Localization and Imaging. Nat. Mater. 2010, 9, 485- 490. 26. Strano, M. S. Nanoscale Radiosurgery. Nat. Mater. 2010, 9, 467. 27. Ge, H.; Riss, P. J.; Mirabello, V.; Calatayud, D. G.; Flower, S. E.; Arrowsmith, R. L.; Fryer, T. D.; Hong, Y.; Sawiak, S.; Jacobs, R. M. J., et al. Behavior of Supramolecular Assemblies of Radiometal-Filled and Fluorescent Carbon Nanocapsules in Vitro and in Vivo. Chem 2017, 3, 437-460. 28. De Munari, S.; Sandoval, S.; Pach, E.; Ballesteros, B.; Tobias, G.; Anthony, D. C.; Davis, B. G. In Vivo Behaviour of Glyco-Nai@Swcnt ‘Nanobottles’. Inorg. Chim. Acta 2019, 495, 118933. 29. Shao, L.; Tobias, G.; Huh, Y.; Green, M. L. H. Reversible Filling of Single Walled Carbon Nanotubes Opened by Alkali Hydroxides. Carbon 2006, 44, 2855-2858. 35 30. Martincic, M.; Vranic, S.; Pach, E.; Ballesteros, B.; Kostarelos, K.; Tobias, G. Non- Cytotoxic Carbon Nanocapsules Synthesized Via One-Pot Filling and End-Closing of Multi- Walled Carbon Nanotubes. Carbon 2019, 141, 782-793. 31. Kierkowicz, M.; González-Domínguez, J. M.; Pach, E.; Sandoval, S.; Ballesteros, B.; Da Ros, T.; Tobias, G. Filling Single-Walled Carbon Nanotubes with Lutetium Chloride: A Sustainable Production of Nanocapsules Free of Nonencapsulated Material. ACS Sustain. Chem. Eng. 2017, 5, 2501-2508. 32. Serpell, C. J.; Rutte, R. N.; Geraki, K.; Pach, E.; Martincic, M.; Kierkowicz, M.; De Munari, S.; Wals, K.; Raj, R.; Ballesteros, B., et al. Carbon Nanotubes Allow Capture of Krypton, Barium and Lead for Multichannel Biological X-Ray Fluorescence Imaging. Nat. Commun. 2016, 7, 13118. 33. Song, G.; Cheng, L.; Chao, Y.; Yang, K.; Liu, Z. Emerging Nanotechnology and Advanced Materials for Cancer Radiation Therapy. Adv. Mater. 2017, 29, 1700996. 34. Goel, S.; England, C. G.; Chen, F.; Cai, W. Positron Emission Tomography and Nanotechnology: A Dynamic Duo for Cancer Theranostics. Adv. Drug Deliv. Rev. 2017, 113, 157-176. 35. Kunz-Schughart, L. A.; Dubrovska, A.; Peitzsch, C.; Ewe, A.; Aigner, A.; Schellenburg, S.; Muders, M. H.; Hampel, S.; Cirillo, G.; Iemma, F., et al. Nanoparticles for Radiooncology: Mission, Vision, Challenges. Biomaterials 2017, 120, 155-184. 36. Ehlerding, E. B.; Grodzinski, P.; Cai, W.; Liu, C. H. Big Potential from Small Agents: Nanoparticles for Imaging-Based Companion Diagnostics. ACS Nano 2018, 12, 2106-2121. 37. Singh, R.; Pantarotto, D.; Lacerda, L.; Pastorin, G.; Klumpp, C.; Prato, M.; Bianco, A.; Kostarelos, K. Tissue Biodistribution and Blood Clearance Rates of Intravenously Administered Carbon Nanotube Radiotracers. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 3357-3362. 38. Wang, J. T.; Cabana, L.; Bourgognon, M.; Kafa, H.; Protti, A.; Venner, K.; Shah, A. M.; Sosabowski, J.; Mather, S. J.; Roig, A., et al. Magnetically Decorated Multi-Walled Carbon Nanotubes as Dual Mri and Spect Contrast Agents. Adv. Funct. Mater. 2014, 24, 1880-1894. 39. McDevitt, M. R.; Chattopadhyay, D.; Jaggi, J. S.; Finn, R. D.; Zanzonico, P. B.; Villa, C.; Rey, D.; Mendenhall, J.; Batt, C. A.; Njardarson, J. T., et al. Pet Imaging of Soluble Yttrium-86- Labeled Carbon Nanotubes in Mice. PLoS ONE 2007, 2, e907. 40. Ruggiero, A.; Villa, C. H.; Holland, J. P.; Sprinkle, S. R.; May, C.; Lewis, J. S.; Scheinberg, D. A.; McDevitt, M. R. Imaging and Treating Tumor Vasculature with Targeted Radiolabeled Carbon Nanotubes. Int. J. Nanomed. 2010, 5, 783-802. 41. Hartman, K. B.; Hamlin, D. K.; Wilbur, D. S.; Wilson, L. J. 211atcl@Us-Tube Nanocapsules: A New Concept in Radiotherapeutic-Agent Design. Small 2007, 3, 1496-1499. 42. Cisneros, B. T.; Law, J. J.; Matson, M. L.; Azhdarinia, A.; Sevick-Muraca, E. M.; Wilson, L. J. Stable Confinement of Positron Emission Tomography and Magnetic Resonance Agents within Carbon Nanotubes for Bimodal Imaging. Nanomedicine 2014, 9, 2499-2509. 43. Matson, M. L.; Villa, C. H.; Ananta, J. S.; Law, J. J.; Scheinberg, D. A.; Wilson, L. J. Encapsulation of Alpha-Particle-Emitting 225ac3+ Ions within Carbon Nanotubes. J. Nucl. Med. 2015, 56, 897-900. 44. Ren, Y.; Pastorin, G. Incorporation of Hexamethylmelamine inside Capped Carbon Nanotubes. Adv. Mater. 2008, 20, 2031-2036. 45. Shao, L.; Lin, T.-W.; Tobias, G.; Green, M. L. H. A Simple Method for the Containment and Purification of Filled Open-Ended Single Wall Carbon Nanotubes Using C60 Molecules. Chem. Commun. 2008, 2164-2166. 36 46. Bayouth, J. E.; Macey, D. J.; Kasi, L. P.; Fossella, F. V. Dosimetry and Toxicity of Samarium-153-Edtmp Administered for Bone Pain Due to Skeletal Metastases. J. Nucl. Med. 1994, 35, 63-9. 47. Yeong, C.-H.; Abdullah, B. J. J.; Ng, K.-H.; Chung, L.-Y.; Goh, K.-L.; Sarji, S. A.; Perkins, A. C. Production and First Use of 153smcl3-Ion Exchange Resin Capsule Formulation for Assessing Gastrointestinal Motility. Appl. Radiat. Isot. 2012, 70, 450-455. 48. Anderson, P. Samarium for Osteoblastic Bone Metastases and Osteosarcoma. Expert Opin. Pharmacother. 2006, 7, 1475-86. 49. Hashikin, N. A.; Yeong, C. H.; Abdullah, B. J.; Ng, K. H.; Chung, L. Y.; Dahalan, R.; Perkins, A. C. Neutron Activated Samarium-153 Microparticles for Transarterial Radioembolization of Liver Tumour with Post-Procedure Imaging Capabilities. PLoS One 2015, 10, e0138106. 50. Fessi, H.; Salim, H.; Barbos, D. Holmium-Loaded Plla Nanoparticles for Intratumoral Radiotherapy Via the Tmt Technique: Preparation, Characterization, and Stability Evaluation after Neutron Irradiation Au - Hamoudeh, Misara. Drug Dev. Ind. Pharm. 2008, 34, 796-806. 51. Bult, W.; Varkevisser, R.; Soulimani, F.; Seevinck, P. R.; de Leeuw, H.; Bakker, C. J. G.; Luijten, P. R.; van het Schip, A. D.; Hennink, W. E.; Nijsen, J. F. W. Holmium Nanoparticles: Preparation and in Vitro Characterization of a New Device for Radioablation of Solid Malignancies. Pharm. Res. 2010, 27, 2205-2212. 52. Di Pasqua, A. J.; Huckle, J. E.; Kim, J.-K.; Chung, Y.; Wang, A. Z.; Jay, M.; Lu, X. Preparation of Neutron-Activatable Holmium Nanoparticles for the Treatment of Ovarian Cancer Metastases. Small 2012, 8, 997-1000. 53. Cagle, D. W.; Thrash, T. P.; Alford, M.; Chibante, L. P. F.; Ehrhardt, G. J.; Wilson, L. J. Synthesis, Characterization, and Neutron Activation of Holmium Metallofullerenes. J. Am. Chem. Soc. 1996, 118, 8043-8047. 54. Cagle, D. W.; Kennel, S. J.; Mirzadeh, S.; Alford, J. M.; Wilson, L. J. In Vivo Studies of Fullerene-Based Materials Using Endohedral Metallofullerene Radiotracers. Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 5182-5187. 55. Di Pasqua, A. J.; Yuan, H.; Chung, Y.; Kim, J. K.; Huckle, J. E.; Li, C.; Sadgrove, M.; Tran, T. H.; Jay, M.; Lu, X. Neutron-Activatable Holmium-Containing Mesoporous Silica Nanoparticles as a Potential Radionuclide Therapeutic Agent for Ovarian Cancer. J. Nucl. Med. 2013, 54, 111-116. 56. Di Pasqua, A. J.; Miller, M. L.; Lu, X.; Peng, L.; Jay, M. Tumor Accumulation of Neutron-Activatable Holmium-Containing Mesoporous Silica Nanoparticles in an Orthotopic Non-Small Cell Lung Cancer Mouse Model. Inorg. Chim. Acta 2012, 393, 334-336. 57. Munaweera, I.; Koneru, B.; Shi, Y.; Pasqua, A. J. D.; Kenneth J. Balkus, J. Chemoradiotherapeutic Wrinkled Mesoporous Silica Nanoparticles for Use in Cancer Therapy. APL Mater. 2014, 2, 113315. 58. Munaweera, I.; Shi, Y.; Koneru, B.; Saez, R.; Aliev, A.; Di Pasqua, A. J.; Balkus, K. J. Chemoradiotherapeutic Magnetic Nanoparticles for Targeted Treatment of Nonsmall Cell Lung Cancer. Mol. Pharm. 2015, 12, 3588-3596. 59. Kim, J.; Luo, Z.-X.; Wu, Y.; Lu, X.; Jay, M. In-Situ Formation of Holmium Oxide in Pores of Mesoporous Carbon Nanoparticles as Substrates for Neutron-Activatable Radiotherapeutics. Carbon 2017, 117, 92-99. 60. Kim, J.; Jay, M. Neutron-Activatable Radionuclide Cancer Therapy Using Graphene Oxide Nanoplatelets. Nucl. Med. Biol. 2017, 52, 42-48. 37 61. Smits, M. L. J.; Nijsen, J. F. W.; van den Bosch, M. A. A. J.; Lam, M. G. E. H.; Vente, M. A. D.; Mali, W. P. T. M.; van het Schip, A. D.; Zonnenberg, B. A. Holmium-166 Radioembolisation in Patients with Unresectable, Chemorefractory Liver Metastases (Hepar Trial): A Phase 1, Dose-Escalation Study. Lancet Oncol. 2012, 13, 1025-1034. 62. Radosa, C. G.; Radosa, J. C.; Grosche-Schlee, S.; Zöphel, K.; Plodeck, V.; Kühn, J. P.; Kotzerke, J.; Hoffmann, R.-T. Holmium-166 Radioembolization in Hepatocellular Carcinoma: Feasibility and Safety of a New Treatment Option in Clinical Practice. Cardiovasc. Interv. Radiol. 2019, 42, 405-412. 63. Kierkowicz, M.; Pach, E.; Santidrián, A.; Sandoval, S.; Gonçalves, G.; Tobías-Rossell, E.; Kalbáč, M.; Ballesteros, B.; Tobias, G. Comparative Study of Shortening and Cutting Strategies of Single-Walled and Multi-Walled Carbon Nanotubes Assessed by Scanning Electron Microscopy. Carbon 2018, 139, 922-932. 64. Manual for Reactor Produced Radioisotopes, Iaea-Tecdoc-1340. Vienna, Austria, 2003. 65. Giulio, C.; Simonelli, F.; Holzwarth, U.; Gibson, P. In Isotopes in Nanoparticles: Fundamentals and Applications, Llop-Roig, J.; Gómez-Vallejo, V.; Gibson, N., Eds. Pan Stanford Publishing Pte. Ltd.: 2016. 66. Martincic, M.; Pach, E.; Ballesteros, B.; Tobias, G. Quantitative Monitoring of the Removal of Non-Encapsulated Material External to Filled Carbon Nanotube Samples. Phys. Chem. Chem. Phys. 2015, 17, 31662-31669. 67. Hutchison, J. L.; Grobert, N.; Zakalyukin, R. M.; Eliseev, A. A.; Chernisheva, M. V.; Kumskov, A. S.; Grigoriev, Y. V.; Krestinin, A. V.; Freitag, B.; Kiselev, N. A. The Behaviour of 1d Cui Crystal@Swnt Nanocomposite under Electron Irradiation. AIP Conference Proceedings 2008, 999, 79-92. 68. Cabana, L.; Ballesteros, B.; Batista, E.; Magén, C.; Arenal, R.; Oró-Solé, J.; Rurali, R.; Tobias, G. Synthesis of Pbi2 Single-Layered Inorganic Nanotubes Encapsulated within Carbon Nanotubes. Adv. Mater. 2014, 26, 2016-2021. 69. Hart, M.; White, E. R.; Chen, J.; McGilvery, C. M.; Pickard, C. J.; Michaelides, A.; Sella, A.; Shaffer, M. S. P.; Salzmann, C. G. Encapsulation and Polymerization of White Phosphorus inside Single-Wall Carbon Nanotubes. Angew. Chem. Int. Ed. 2017, 56, 8144-8148. 70. Sandoval, S.; Kepić, D.; Pérez del Pino, Á.; György, E.; Gómez, A.; Pfannmoeller, M.; Tendeloo, G. V.; Ballesteros, B.; Tobias, G. Selective Laser-Assisted Synthesis of Tubular Van Der Waals Heterostructures of Single-Layered Pbi2 within Carbon Nanotubes Exhibiting Carrier Photogeneration. ACS Nano 2018, 12, 6648-6656. 71. Vasylenko, A.; Marks, S.; Wynn, J. M.; Medeiros, P. V. C.; Ramasse, Q. M.; Morris, A. J.; Sloan, J.; Quigley, D. Electronic Structure Control of Sub-Nanometer 1d Snte Via Nanostructuring within Single-Walled Carbon Nanotubes. ACS Nano 2018, 12, 6023-6031. 72. Slade, C. A.; Sanchez, A. M.; Sloan, J. Unprecedented New Crystalline Forms of Snse in Narrow to Medium Diameter Carbon Nanotubes. Nano Lett. 2019, 19, 2979-2984. 73. Tse, J. W.; Wiebe, L. I.; Noujaim, A. A. High Specific Activity [Samarium-153] Edta for Imaging of Experimental Tumor Models. J. Nucl. Med. 1989, 30, 202-208. 74. Fidler, I. J. Selection of Successive Tumour Lines for Metastasis. Nat. New Biol. 1973, 242, 148-9. 75. Siddikuzzaman; Grace, V. M. B. Inhibition of Metastatic Lung Cancer in C57bl/6 Mice by Liposome Encapsulated All Trans Retinoic Acid (Atra). Int. Immunopharmacol. 2012, 14, 570-579. 38 76. Perez Ruiz de Garibay, A.; Spinato, C.; Klippstein, R.; Bourgognon, M.; Martincic, M.; Pach, E.; Ballesteros, B.; Ménard-Moyon, C.; Al-Jamal, K. T.; Tobias, G., et al. Evaluation of the Immunological Profile of Antibody-Functionalized Metal-Filled Single-Walled Carbon Nanocapsules for Targeted Radiotherapy. Sci. Rep. 2017, 7, 42605. 77. Kallini, J. R.; Gabr, A.; Thorlund, K.; Balijepalli, C.; Ayres, D.; Kanters, S.; Ebrahim, S.; Mills, E.; Lewandowski, R. J.; Salem, R. Comparison of the Adverse Event Profile of Therasphere® with Sir-Spheres® for the Treatment of Unresectable Hepatocellular Carcinoma: A Systematic Review. Cardiovasc. Interv. Radiol. 2017, 40, 1033-1043. 78. Package Insert - TheraSphere® Yttrium-90 Glass Microspheres. 79. Package Insert - SIR-Spheres® Y-90 resin microspheres. 80. FDA Humanitarian Device Exemption: H980006 Summary of safety and portable benefit, 1999. 81. Tobias, G.; Shao, L.; Salzmann, C. G.; Huh, Y.; Green, M. L. H. Purification and Opening of Carbon Nanotubes Using Steam. The Journal of Physical Chemistry B 2006, 110, 22318-22322. 82. Martincic, M.; Frontera, C.; Pach, E.; Ballesteros, B.; Tobias, G. Synthesis of Dry Smcl3 from Sm2o3 Revisited. Implications for the Encapsulation of Samarium Compounds into Carbon Nanotubes. Polyhedron 2016, 116, 116-121. 83. Sandoval, S.; Kierkowicz, M.; Pach, E.; Ballesteros, B.; Tobias, G. Determination of the Length of Single-Walled Carbon Nanotubes by Scanning Electron Microscopy. MethodsX 2018, 5, 1465-1472. 84. Šefl, M.; Kyriakou, I.; Emfietzoglou, D. Technical Note: Impact of Cell Repopulation and Radionuclide Uptake Phase on Cell Survival. Med. Phys. 2016, 43, 2715-2720. 85. Stabin, M. G.; Konijnenberg, M. W. Re-Evaluation of Absorbed Fractions for Photons and Electrons in Spheres of Various Sizes. J. Nucl. Med. 2000, 41, 149-60. 86. Schiller, U.; Hoffmann, W.; Bamberg, M.; Rodemann, H. P. All-Trans-Retinoic Acid Modulates the Radiosensitivity of Proliferating Cells. Int. J. Cancer 1997, 70, 488-90. 87. Chapman, J. N. N., A.E. Radiotherapy Treatment Planning: Linear-Quadratic Radiobiology. CRC Press: 2015. 88. Dale, R. G. Dose-Rate Effects in Targeted Radiotherapy. Phys. Med. Biol. 1996, 41, 1871-84. TABLE OF CONTENTS 39 40 Supporting Information Neutron Activated 153Sm Sealed in Carbon Nanocapsules for In Vivo Imaging and Tumor Radiotherapy Julie T.-W. Wanga, Rebecca Klippsteina, Markus Martincicb, Elzbieta Pachc, Robert Feldmand, Martin Sefle,f, Yves Micheld, Daniel Askera, Jane K. Sosabowskig, Martin Kalbach, Tatiana Da Rosi, Cécilia Ménard-Moyonj, Alberto Biancoj, Ioanna Kyriakoue, Dimitris Emfietzogloue, Jean- Claude Saccavinid, Belén Ballesterosc, Khuloud T. Al-Jamala,* and Gerard Tobiasb,* a Institute of Pharmaceutical Science, King’s College London, London SE1 9NH, UK b Institut de Ciència de Materials de Barcelona (ICMAB-CSIC), Campus UAB, 08193 Bellaterra, Barcelona, Spain c Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC and the Barcelona Institute of Science and Technology, Campus UAB, 08193 Bellaterra, Barcelona, Spain d Cis Bio International Ion Beam Applications SA, Gif sur Yvette 91192, France e Medical Physics Laboratory, University of Ioannina Medical School, Ioannina GR-45110, Greece 1 f Faculty of Nuclear Sciences and Physical Engineering, Czech Technical University in Prague, Prague 11519, Czech Republic g Centre for Molecular Oncology, Barts Cancer Institute, Queen Mary University of London, London EC1M 6BQ, UK h J. Heyrovsky Institute of the Physical Chemistry, Dolejskova 3, 182 23 Prague 8, Czech Republic i INSTM Unit of Trieste, Department of Chemical and Pharmaceutical Sciences, University of Trieste, Via L. Giorgieri 1, 34127 Trieste, Italy j University of Strasbourg, CNRS, Immunology, Immunopathology and Therapeutic Chemistry, UPR 3572, 67000 Strasbourg, France. *Corresponding authors E-mail: gerard.tobias@icmab.es, khuloud.al-jamal@kcl.ac.uk, belen.ballesteros@icn2.cat. 2 Figure S1. Schematic representation of the strategies employed for the hermetic sealing of radionuclides into the cavities of carbon nanotubes (hot nanocapsules). a) Previously employed strategies where radionuclides (125I and 64Cu) are directly filled into carbon nanotubes[1]. b) Present strategy, also depicted in Fig. 1a (main text), where a non-radioactive enriched precursor is initially encapsulated (152Sm) and it is activated into its radioactive form by neutron irradiation (153Sm) in the last step. In both schematic representations, the wavy lines indicate radioactivity emerging from radionuclides (highlighted in the text in red). 3 Figure S2. Length distribution of 152Sm@CNT samples. a) Length distribution histogram and b) box plot analysis of SWNT; c) Length distribution histogram and d) box plot analysis of MWNT. Table S1. Descriptive analysis of the length distribution of 152Sm@CNT samples. Lower Q1 Q3 Maximum N Lower Maximum Median adjacent 25 th 75 th adjacent Number of observation observation measured CNT (nm) observation percentile percentile observation (nm) (nm) (nm) (nm) (nm) (nm) SWNT 200 199 41 41 120 397 794 3500 MWNT 200 331 40 40 192 495 932 4682 4 Figure S3. Diameter distribution of 152Sm@CNT samples. a) Diameter distribution histogram and b) box plot analysis of SWNT; c) Diameter distribution histogram and d) box plot analysis of MWNT. Table S2. Descriptive analysis of the diameter distribution of 152Sm@CNT samples. Lower Q1 Q3 Maximum N Lower Maximum Median adjacent 25 th 75 th adjacent Number of observation observation measured CNT (nm) observation percentile percentile observation (nm) (nm) (nm) (nm) (nm) (nm) SWNT 200 2.1 0.5 0.5 1.6 3.2 5.5 6.4 MWNT 200 10.7 2.5 3.4 8.8 12.8 18.2 26.2 5 Figure S4. ITLC analysis of 153Sm@SWNT and 153Sm@MWNT following neutron irradiation of 152Sm@SWNT and 152Sm@MWNT respectively. Signals at the application point represent the stably encapsulated radionuclides. Figure S5. HRTEM images of Sm-filled SWNTs before (‘Cold’ nanocapsules) and after (‘Hot’ nanocapsules) irradiation. 6 Figure S6. HAADF STEM images of a 152SmCl3@SWNT showing that during electron microscopy observation the filling material displaces inside the SWNT. a) Initially acquired image and b) an image of the same area acquired after ca. 30 seconds. Displacement of the filling material is indicated with white bars. 7 Table S3. Neutron activated isotopes (of clinical interest) loaded onto nanocarriers. Nanocarrier Stable Neutron Neutron Reported Specific Ref. isotope irradiation flux radioactivitya radioactivityb time (n∙cm‐2∙s‐1) (MBq/mg) 152 MWNT Sm 96 h 1.6x1014 11.37 GBq/mg 11370.0 This work 152 SWNT Sm 96 h 1.6x1014 6.33 GBq/mg 6330.0 This work C82 165 Ho 1 min 4.3x1014 16.13 Ci/mg 0.6 [2] MCN 165 Ho 10 h 5.5x1012 208 Ci/100 g 77.0 [3] HoIG 165 Ho 0.45 h 7.0x1012 250 Ci/mg 9.3 [4] Silica NPs 165 Ho 1h 3.5x1012 213.6 Ci/mg 7.9 [5] 165 Silica NPs Ho 2.2 hc 5.5x1012 129 MBq/10.7 mg 12.1 [6] Silica NPs 165 Ho 2h 5.5x1012 150 Ci/mg 5.6 [7] Silica NPs 165 Ho 3h 7.7x1012 300 Ci/mg 11.1 [7] 165 AcAc Ho 13 min 5.5x1012 0.25 mCi/10 mg 0.9 [8] 165 AcAc Ho 1h 5.0×1012 600 MBq/50 mg 12.0 [9] 165 PLLA Ho 1h 1.1×1013 27.4 GBq/g 27.4 [10] MWNT: Multi‐walled carbon nanotubes; SWNT: Single‐walled carbon nanotubes; C82: Endohedral Metallofulerenes; HoIG: Holmium iron garnet; NPs: Nanoparticles; MCN: Mesoporous carbon nanoparticles; AcAc: NPs from acetylacetonate; PLLA: Poly‐L‐lactide NOTES: aValues as reported in the reference provided b Specific radioactivities calculated from “Reported activity” c Irradiation up to 18h was performed but the resulting activity is not reported 8 153 153 Sm@SWNT Sm@MWNT 80 80 70 70 60 60 % ID/organ % ID/organ 50 50 40 4h 40 4h 30 24 h 30 24 h 20 20 10 10 0 0 Figure S7. Tissue biodistribution of 153Sm@SWNT and 153Sm@MWNT (presented as %ID per organ). C57/Bl6 mice were i.v. injected with 200 µg of 153Sm@SWNT or 153Sm@MWNT (~1 MBq). The radioactivity of major organs sampled at specified time points were measured by γ-scintigraphy. The results are expressed as % ID/organ and presented as mean ± S.D. (n=3-4). Figure S8. Live bioluminescence images of untreated mice or mice treated with 153 Sm@SWNT or 153Sm@MWNT. B16F10-Luc tumor-bearing mice were i.v. injected with 20 MBq in 200 µg of 153Sm@SWNT or 153Sm@MWNT on day 8 post-tumor inoculation. Bioluminescence signals correspond to metabolic activity of luciferase-expressing B16F10 cells in the lung. 9 Figure S9. Histological examination of major organs from C57BL/6 mice at 24 h post injection of 153Sm@SWNT and 153Sm@MWNT. C57BL/6 mice were i.v. injected with 200 µg of 153Sm@SWNT or 153Sm@MWNT (~1 MBq). At 24 h post injection, heart, lung, kidney, liver, and spleen were excised and formalin-fixed, stained with H&E (for tissue necrosis) or Neutral Red (for CNT visualization). As a guide to the eye arrows point to some CNT aggregates. Scale bars: 50 μm. 10 Figure S10. Fit of the experimental biodistribution data (% ID) in the lung using equation 3 1 1 described in the manuscript. Fitted parameters are ID0  16.6 % , k  0.0162 h m  2.32 h for 1 1 SWNT, and ID0  28.0 % , k  0.0160 h m  0.826 h for MWNT. Experimental data are presented by circle or square dots. Lines (solid or dotted) are the fitted curve calculated using equation 3. 11 Experimental Details Materials and Reagents: Chemical vapor deposition (CVD) Elicarb® SWNTs and MWNTs were supplied by Thomas Swan & Co. Ltd (UK) as a solid powder. The SWNT material also contains 152 a fraction of double-walled carbon nanotubes. Enriched Sm2O3 were provided by CIS-Bio International-Ion Beam Applications (France). Hydrogen peroxide, sulphuric acid and nitric acids were purchased from Panreac AppliChem (Spain) for CNT pre-treatment. Instant thin layer chromatography paper impregnated with silica gel (ITLC-SG) was obtained from Agilent Technologies (UK). Isoflurane for anesthesia was purchased from Abbott (IsoFlo®, Abbott Laboratorie Ltd, UK). Purification and Shortening of CNTs: Both SWNTs and MWNTs were initially treated to shorten the tubes, open their ends and remove carbonaceous and metallic (catalyst) impurities. SWNTs were exposed to a combined piranha-steam treatment, whereas MWNTs underwent a combined H2SO4:HNO3-steam treatment following previously reported protocols.[11] Synthesis of 152SmCl3 from 152Sm2O3: Enriched 152Sm2O3 was transformed to anhydrous 152SmCl3 following the protocol reported for the synthesis of anhydrous SmCl3 with natural isotopic distribution (non-enriched).[12] The synthesis of the anhydrous metal halide was performed by dissolution of the enriched metal oxide in HCl, followed by dehydration of the collected solid at 152 240 °C under dynamic vacuum. The synthesized anhydrous SmCl3 is highly hygroscopic and was handled under an inert atmosphere. 152 Filling of CNTs with SmCl3: Cut and purified SWNT (100 mg) or MWNT (200 mg) and 152 SmCl3 were ground together in a weight ratio 1:10 (CNTs:152SmCl3). The materials were ground using an agate mortar and pestle inside an argon filled glove box until the mixture presented a uniform color. The samples were split in smaller fractions, placed inside silica tubes and sealed under vacuum. The resulting silica ampoules were placed inside a horizontal tubular furnace and annealed for 12 h at 900 °C (SWNT) or 1200 °C (MWNT) thus leading to the formation of carbon nanocapsules (closed-ended filled CNTs). The samples were recovered from the ampoules and the non-encapsulated material, external to the CNTs was dissolved and washed away. The removal of the external material was followed by UV-Vis spectroscopy of the 12 152 collected filtrates, until no more SmCl3 was detected in the washings.[13] Initially the filled nanotubes were soaked in 200 mL of water containing 0.6 M HCl. The sample was then collected by filtration over a 0.2 μm polycarbonate membrane. This “pre-washing” step was followed by washing the sample three times in 200 mL acidic water at 80 °C for 24 h under constant stirring each time. The sample was collected by filtration between washings. A final washing step was performed using 200 mL of pure water, without the addition of HCl, whilst keeping the rest parameters constant. The collected solid powder was dried at 80 °C overnight. 152 Neutron activation of Sm-filled CNTs: Vacuum sealed silica ampoules containing 30 mg of 152 152 either Sm@SWNT or Sm@MWNT were placed in an aluminum capsule, weighed down with a lead-weight in a pool-type reactor (OSIRIS, CEA Saclay, France) and irradiated at a neutron flux of 1.6x1014 n·cm-2·s-1 for 96 h. The irradiation protocol was established according to equation (1):[14] . σΦ λ A 1 (1) where A is the predicted activity of the radioisotope produced (Bq g-1), M is the atomic mass of the target element (152 g mol-1 for 152Sm), Φ is the neutron flux of the reactor (1.6x1014 n cm-2s- 1 ), σ is the thermal neutron activation cross-section of the target isotope (206 barns for 153Sm), λ is the decay constant (0.693/T1/2), (T1/2 is the half life of the target isotope, which is 46.27 h for 153 Sm) and t is the irradiation time (96 h). The value of A for the given conditions can then be used to make a prediction of the nanotubes, by factoring in the mass of nanotubes and the corresponding percentage content of 152Sm that was measured before the irradiation. After removing the ampoules from the pool, they were allowed to cool down, removed from the aluminum casing, and then transferred to CIS bio International, Saclay (France) where they were processed in a fully sealed radioprotection hot cell. Radioactivity was measured using a dose calibrator (VDC 404, Veenstra Instruments, The Netherlands). The ampoules were then shattered and the contents (153Sm@SWNT and 153 Sm@MWNT powders) were separately suspended in a volume of 1% Pluronic® F-127 saline (0.9 % NaCl) solution, using sonication such to produce a suspension with a loading of 0.5 mg/mL. The glass fragments from the broken silica ampoule were separated from the bulk of the suspension by sedimentation. Large aggregates of 153 Sm@CNTs were also precluded in this way. In order to transfer this product to the in vivo 13 facilities in vials, keeping within their local limits of activity handling, only a fraction of this suspension was then separated into a final vial, which was then shipped and arrived in a few days. On arrival, a sonication and a series of dilution steps were performed to ensure that the final dispersion injected in mice was physically stable and free of aggregates, as detailed in the in vivo section. 152 Quantification of SmCl3 filling yields by ICP-MS: As a analytical validation measure, each carbon nanotube sample was subdivided into fractions of different sizes in the range of 1-15 mg, and was weighed three times on a pre-calibrated analytical balance. The samples were then added to a PTFE-TFM reactor in an Anton Paar Multiwave 3000, fitted with an 8SXF100 rotor. 6 mL 65% Suprapur® nitric acid (Merck KGaA, Germany), and then 2 mL of 30% Suprapur® hydrogen peroxide (Merck KGaA, Germany) were dispensed from calibrated micropipettes and the microwave digestion was completed by applying 800 W for 40 minutes, with an initial ramp of 5 min. The contents of each reactor were then transferred to a polyproylene flask, followed by a series of washings with de-ionised water (ELGA Labwater PURELAB® Classic water purifying system; resistivity 18.2 MΩ/cm), and the volume was then made up to 25 mL. Two additional control solutions were prepared and analyzed. It was demonstrated that no impurities were introduced during this process using a “preparation blank,” consisting of all of the reagents except the nanotubes, that was passed through the microwave at the same time as the actual samples. It was also confirmed that all of the material that was added to the reactors was recovered by preparing a solution from the final washing. Prior to analysis by ICP-MS, all of the samples were filtered with a 33 mm 0.22 µm MILLEX ® GV PVDT filter. ICP-MS analysis was carried out at Cis Bio International using an ICP-MS with a quadrupole collision cell (PerkinElmer Sci EX ELAN® DRC II). The system check consisted of: a daily performance report using a standard reference solution of magnesium, indium, uranium, cerium and barium, with relative standard deviations (RSD) within 0.4 to 1.1% of the anticipated value. The sample sequence then consisted of triplicate injections of: a blank containing de-ionized water, the “preparation blank” and then the actual samples, separated by injections of de-ionized 152 water. The quantification was performed using an external Sm calibration standard. The stability of the measurements was monitored by spiking each sample with an identical level of a rhodium standard of 5 ppb and this gave RSDs of less than 1%. The range of concentrations 14 152 chosen for the Sm calibration depended on the particular sample and covered the ranges of either 1 to 5 ppb or 1 to 30 ppb, and all of Pearson correlation coefficients R2 coefficients were greater than 0.9999. The results corresponding to the overall % 152Sm were then reported to one decimal place. Electron microscopy: HAADF-STEM images were acquired at 20 kV on an FEI Magellan XHR Scanning Electron Microscope (SEM) with the use of a specially adapted detector. HRTEM images were acquired on a FEI Tecnai G2 F20 microscope at 200 kV. Samples were dispersed in ethanol and deposited onto lacey carbon Cu grids. 153Sm@CNT were imaged after the complete decay of radioactivity. EDX was carried out on a FEI Quanta SEM microscope equipped with an EDAX detector at 20 kV. Animals: All in vivo experiments were conducted under the authority of project and personal licenses granted by the UK Home Office and the UKCCCR Guidelines (1998). Female C57BL/6 mice aged 6-8 weeks were purchased from Harlan (UK) and used for all in vivo studies. 153 ITLC examination of Sm@CNTs: ITLC was performed to examine whether there was free 153 153 152 Sm present in the Sm@CNT suspensions (neutron irradiated Sm@CNT) prior to 153 153 injection. Aliquots of Sm@SWNT and Sm@MWNT were spotted on TLC strips and then developed in 0.1 M ammonium acetate containing 50 mM EDTA as a mobile phase. Strips were allowed to dry and counted quantitatively using a cyclone phosphor detector (Packard Biosciences, UK). 153 153 Preparation of Sm-CNTs for in vivo studies: The Sm@CNT suspensions underwent different dilutions to contain the appropriate radioactivity for different in vivo studies. The 153 dilution was carried out by mixing the Sm@CNT dispersions with different amounts of non- 152 irradiated Sm@CNT suspensions (2 mg/mL in 1% Pluronic® F-127 saline). The resulting mixture was then sonicated for 10-15 min. Each injection dose, either for imaging, biodistribution or therapy studies, contained the same amount of CNT (200 µg), but possessed different radioactivity which are specified in their experimental sections accordingly. Injection suspensions were added with 0.1 M EDTA (one twentieth of the injection volume) to chelate any free 153Sm. 15 153 In vivo SPECT/CT imaging of Sm@CNT following i.v. injection: The biodistribution of 153 Sm@CNTs was firstly assessed by 3D whole body SPECT/CT imaging. Mice were anaesthetized by isoflurane inhalation during imaging. Each injection dose for 153Sm@SWNT or 153 Sm@MWNT suspensions contained approximately 10 MBq. Injection suspensions were 153 added with 0.1 M EDTA (one twentieth of the injection volume) to chelate free Sm. Immediately after injection, and at 4 h and 24 h, mice were imaged using the Nano-SPECT/CT scanner (Bioscan, USA). SPECT images of each mouse were taken in 24 projections over 30-40 min using a four-head scanner with 1.4 mm pinhole collimators. CT scans were performed at the end of each SPECT acquisition. All images were reconstructed by MEDISO software (Medical Imaging Systems), and SPECT and CT images were merged using the InVivoScopeTM software (Bioscan, USA). 153 Pharmacokinetics and organ biodistribution of Sm@CNTs by γ-scintigraphy: Mice 153 153 anaesthetized by isoflurane inhalation were injected with Sm@SWNT or Sm@MWNT suspensions via a tail vein. Each injection dose of 153Sm@SWNT or 153Sm@MWNT suspensions contained approximately 0.8 MBq. Blood was sampled from a tail vein at 5 min, 30 min, 4 h or 24 h after injection. At 4 h and 24 h, mice were sacrificed and major organs were excised and weighed. To assess the excretion profiles, animals were housed individually in metabolic cages where the mice have free access to water but not food. Urine and feces were collected over 24 h after injection. The radioactivity of each sample (i.e. tissues, blood, urine or feces) was measured by γ-scintigraphy (LKB Wallac 1282 Compugamma, PerkinElmer) and the results were expressed as percentage injection dose per sample (% ID) or per g of sample (% ID/g). Collected organs were fixed in 10% neutral buffered formalin for histological examination. B16F10-Luc lung metastasis tumor model: Mice anaesthetized by isoflurane inhalation were injected with 5 x 105 B16F10-Luc cells in 0.2 mL of PBS via a tail vein to establish pulmonary melanoma metastases. Following tumor inoculation, in vivo quantitative bioluminescence imaging was performed on day 7, 10, 13 and 16 to monitor the tumor growth (IVIS Lumina III, Perkin-Elmer, UK). Mice under anesthesia were subcutaneously injected with luciferin (D- luciferin potassium salt, Perkin-Elmer, UK) at 150 mg/kg and imaged 10 min after injection. Bioluminescence signals from regions of interests were measured using Living Image software (Perkin-Elmer, UK) and recorded as total flux (photons/sec). 16 153 In vivo Sm@CNT radiotherapy studies: To determine the radiotherapeutic action of 153 153 Sm@SWNT and Sm@MWNT, B16F10-Luc tumor bearing C57BL/6 mice were randomly divided into three groups (n = 10): untreated, 153Sm@SWNT and 153Sm@MWNT. On day 8 post 153 tumor inoculation, mice were intravenously injected with 200 µg of Sm@SWNT or 153 Sm@MWNT suspensions containing approximately 20 MBq. Mice were sacrificed on day 16 post tumor inoculation, and major organs including lung, liver, spleen, heart, and kidney were weighed and fixed in 10% neutral buffered formalin for histological examination. Tumor growth was monitored by bioluminescence imaging as described previously. The tumor growth data was expressed as mean ± SEM (standard error of the mean), with n denoting the number of animals. Significant differences were examined using one-way ANOVA followed by Tukey's multiple comparison test. *: p < 0.05; ***: p < 0.001 Histological examination: Major organs excised from the mice in the biodistribution study (i.e. 153 24 h post injection) and from the mice after Sm@CNT radiotherapy were subjected to histological examination. Harvested fixed organs were paraffin-embedded and sectioned for haematoxylin and eosin (H&E) or Neutral Red staining according to standard histological protocols at the Royal Veterinary College (UK). All stained sections were analysed using a Leica DM 1000 LED Microscope (Leica Microsystems, UK) coupled with a CCD digital camera (Qimaging, UK). References [1] a) S. Y. Hong, G. Tobias, K. T. Al-Jamal, B. Ballesteros, H. Ali-Boucetta, S. Lozano-Perez, P. D. Nellist, R. B. Sim, C. Finucane, S. J. Mather, M. L. H. Green, K. Kostarelos, B. G. Davis, Nat. Mater. 2010, 9, 485; b) H. Ge, P. J. Riss, V. Mirabello, D. G. Calatayud, S. E. Flower, R. L. Arrowsmith, T. D. Fryer, Y. Hong, S. Sawiak, R. M. J. Jacobs, S. W. Botchway, R. M. Tyrrell, T. D. James, J. S. Fossey, J. R. Dilworth, F. I. Aigbirhio, S. I. Pascu, Chem 2017, 3, 437. [2] D. W. Cagle, S. J. Kennel, S. Mirzadeh, J. M. Alford, L. J. Wilson, Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 5182. [3] J. Kim, Z.-X. Luo, Y. Wu, X. Lu, M. Jay, Carbon 2017, 117, 92. [4] I. Munaweera, Y. Shi, B. Koneru, R. Saez, A. Aliev, A. J. Di Pasqua, K. J. Balkus, Mol. Pharm. 2015, 12, 3588. [5] I. Munaweera, B. Koneru, Y. Shi, A. J. D. Pasqua, J. Kenneth J. Balkus, APL Mater. 2014, 2, 113315. [6] A. J. Di Pasqua, H. Yuan, Y. Chung, J. K. Kim, J. E. Huckle, C. Li, M. Sadgrove, T. H. Tran, M. Jay, X. Lu, J. Nucl. Med. 2013, 54, 111. [7] A. J. Di Pasqua, M. L. Miller, X. Lu, L. Peng, M. Jay, Inorg. Chim. Acta 2012, 393, 334. 17 [8] A. J. Di Pasqua, J. E. Huckle, J.-K. Kim, Y. Chung, A. Z. Wang, M. Jay, X. Lu, Small 2012, 8, 997. [9] W. Bult, R. Varkevisser, F. Soulimani, P. R. Seevinck, H. de Leeuw, C. J. G. Bakker, P. R. Luijten, A. D. van het Schip, W. E. Hennink, J. F. W. Nijsen, Pharm. Res. 2010, 27, 2205. [10] H. Fessi, H. Salim, D. Barbos, Drug Dev. Ind. Pharm. 2008, 34, 796. [11] M. Kierkowicz, E. Pach, A. Santidrián, S. Sandoval, G. Gonçalves, E. Tobías-Rossell, M. Kalbáč, B. Ballesteros, G. Tobias, Carbon 2018, 139, 922. [12] M. Martincic, C. Frontera, E. Pach, B. Ballesteros, G. Tobias, Polyhedron 2016, 116, 116. [13] M. Martincic, E. Pach, B. Ballesteros, G. Tobias, Phys. Chem. Chem. Phys. 2015, 17, 31662. [14] Manual for reactor produced radioisotopes, IAEA-TECDOC-1340, Vienna, Austria 2003. 18