Tellurium-118 as a novel radionuclide for long-term positron emission tomography

Abstract

Positron emission tomography (PET) is a powerful imaging modality that tracks positron-emitting radionuclides, enabling non-invasive and high-resolution visualization of biological targets and processes. However, conventional PET tracers possess short half-lives, which limits their utility for long-term pharmacokinetic studies of slow-clearing agents such as antibodies. To address this unmet need, we developed and evaluated tellurium-118 (¹¹⁸Te), a long-lived radionuclide (T₁/₂ = 6.0 days) that decays via electron capture to produce antimony-118 (¹¹⁸Sb), a short-lived positron emitter (T₁/₂ = 3.5 min). This decay cascade allows ¹¹⁸Te to function as an in vivo radionuclide generator, providing sustained positron emission from a single administration. We successfully produced ¹¹⁸Te and performed PET imaging with NEMA and Derenzo phantoms, confirming its PET images with sufficient spatial resolution. In vivo PET/CT imaging of mice following intravenous administration of ¹¹⁸Te revealed the persistence of PET signals in the abdominal region up to three weeks post-injection. Ex vivo biodistribution analysis confirmed the prolonged retention of radioactivity in abdominal organs, particularly the liver and spleen. These findings demonstrate the feasibility of ¹¹⁸Te as a long-lived PET radionuclide for extended molecular imaging. Its unique decay profile and sustained signal generation make it a promising candidate for longitudinal pharmacokinetic studies and radiotheranostic applications.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-46505-x.

Introduction

Positron emission tomography (PET) imaging has become an indispensable modality for the non-invasive assessment of diverse physiological and pathological conditions. PET imaging for cancer diagnosis often uses carrier agents, such as peptide ligands (e.g., PSMA-11, PSMA-617) or antibodies (e.g., trastuzumab) labeled with positron-emitting radionuclides, to target cancer-specific antigens on cell membranes^1,2. Furthermore, by conjugating carriers with therapeutic radionuclides that emit cytotoxic α- or β⁻particles, targeted delivery of radiation to tumor sites becomes possible. This therapeutic approach is known as the targeted radionuclide therapy (TRT)^3–5. The clinical success of TRT is influenced by several factors, including sufficient expression of target antigens on tumor cells and limited expression of these antigens in normal tissues to minimize off-target radiation effects. PET imaging plays an essential role in assessing both aspects⁶, offering spatial quantification for pre-therapeutic evaluation as well as longitudinal monitoring of therapeutic efficacy.

While TRT agents such as monoclonal antibodies exhibit long biological half-lives due to slow systemic clearance⁷, most conventional PET radionuclides, such as ¹¹C, ¹³N, ¹⁵O, ¹⁸F, ⁶⁸Ga, and ⁶⁴Cu, have relatively short physical half-lives. This limits their capacity to track biodistribution over extended periods, thereby reducing their utility in accurately predicting the pharmacokinetics of long-circulating therapeutic agents⁸. To address this limitation, PET tracers with extended half-lives have been explored. Long-lived PET radionuclides such as ⁸⁹Zr (T₁/₂ = 3.3 days, β⁺ = 22.7%, E₍β ave₎ = 396 keV) and ⁵²Mn (T₁/₂ = 5.6 days, β⁺ = 29.6%, E₍β ave₎ = 242 keV) have gained attention, with ongoing studies applying these tracers to label carrier agents for biodistribution analyses^9–12.

However, most positron-emitting radionuclides with relatively long half-lives ranging from several hours to days decay through processes other than positron emission. For example, many relatively long-lived radionuclides emit unwanted gamma rays with energies higher than 511 keV, increasing background noise. For instance, ⁸⁹Zr emits 909 keV γ-rays, and ⁵²Mn emits 1.434 MeV γ-rays, both of which contribute to scattering and background noise in PET imaging¹³. Although PET systems can be optimized by narrowing the energy window around 511 keV to suppress direct detection of such high-energy photons, scattered photons originating from these emissions may lose energy through Compton interactions within the body or detector materials and subsequently fall within the 511 keV detection window¹⁴. These misclassified events can degrade image quality and quantification accuracy. This issue is expected to be more pronounced in clinical PET imaging, where the larger body volume increases the probability of scatter events, compared to small-animal PET systems.

Extending the tracking period is essential to evaluate slow-clearing carrier agents like therapeutic antibodies. To achieve this imaging capability, we investigated tellurium-118 (¹¹⁸Te), a radionuclide with a physical half-life of 6.0 days. ¹¹⁸Te decays via electron capture into antimony-118 (¹¹⁸Sb), subsequently emitting positrons (β⁺ = 73.5%, Eβ⁺_max = 2635 keV, Eβ⁺_mean = 1186 keV, and EC = 26.5%) with a half-life of 3.5 min to yield stable isotope ¹¹⁸Sn. This decay sequence effectively generates a positron in vivo over a prolonged period from a single administration of ¹¹⁸Te. Additionally, none of these reactions emit unwanted high energy gamma rays that could contribute to background noise.

In this study, we aim to evaluate the feasibility of ¹¹⁸Te for long-term PET imaging and to establish a technological foundation for future development of carrier agents labeled with this novel radionuclide.

Results

Radionuclide production and decay scheme

¹¹⁸Te was produced via the ¹¹⁶Sn(⁴He,2n)¹¹⁸Te nuclear reaction by irradiating a highly enriched metallic ¹¹⁶Sn target (97.8%) with ⁴He ion beams accelerated to 27.9 MeV using an Azimuthally Varying Field (AVF) cyclotron. The product was then dissolved in aqueous HCl and formulated as a solution for subsequent use.

¹¹⁸Te (T₁/₂ = 6.02 days) decays exclusively via electron capture (EC = 100%) to yield radionuclide ¹¹⁸Sb (T₁/₂ = 3.61 min), which further undergoes positron emission (β⁺ = 73.5%, Eβ⁺_max = 2635 keV, Eβ⁺_mean = 1186 keV, and EC = 26.5%) and transitions to stable isotope ¹¹⁸Sn (Fig. 1). The γ-ray spectrum of the purified ¹¹⁸Te solution, acquired using a germanium semiconductor detector, contained prominent peaks corresponding to ¹¹⁸Sb in radioactive equilibrium with ¹¹⁸Te (Supplementary Fig. S1).

Fig. 1.

Decay pattern of ¹¹⁸Te. All nuclear data are from the NuDat 3.0 database³⁶.

Spatial resolution analysis using phantoms

NEMA and Derenzo phantoms were employed to qualitatively evaluate the spatial resolution of PET images obtained from annihilation photons derived by ¹¹⁸Sb generated from ¹¹⁸Te. Cavities with diameters ranging from 6.0 mm to 2.0 mm in the NEMA phantom were resolvable but spatially diffused (Fig. 2a). This diffusion may be partly attributed to the physical properties of positron emission. Conventional PET radionuclides such as F‑18 (E₍β⁺ max₎ = 630 keV, E₍β⁺ mean₎ = 250 keV) and Cu‑64 (E₍β⁺ max₎ = 653 keV, E₍β⁺ mean₎ = 278 keV) emit positrons with relatively low energies. In contrast, ¹¹⁸Sb, generated from ¹¹⁸Te decay, emits high-energy positrons (E₍β⁺ max₎ = 2635 keV, E₍β⁺ mean₎ = 1136 keV; Fig. 1). These high-energy positrons travel longer distances before annihilation, increasing the spatial mismatch between the radionuclide location and the actual annihilation site, which likely contributes to spatial diffusion of imaging.

Fig. 2.

Analysis of spatial resolution provided by ¹¹⁸Te using PET phantoms. (a) PET image of a NEMA phantom loaded with 1.0 MBq of ¹¹⁸Te acquired for 3 h. Acquired raw data were reconstructed using the software implemented in the Clairvivo-PET system (Shimadzu, Japan) and finally analyzed using DICOM viewer Osirix (Pixmeo, Switzerland). Red circles represent the actual size of the cavities of the phantom. (b) PET image of a Derenzo phantom filled with 1.0 MBq ¹¹⁸Te acquired for 3 h using the same method as for the NEMA phantom. (c) Line profiles extracted across the 4.3 mm (red line) and 3.5 mm (blue line) cavities in (b) to assess spatial resolution. ImageJ was used to quantitatively evaluate the degree of signal separation between adjacent cavities. In addition, full-width at half-maximum (FWHM) values were calculated from three distinct signal peaks observed in the 3.5 mm cavities, as indicated by blue arrows.

Also, cavities with diameters of 4.3 mm and 3.5 mm in the Derenzo phantom were both resolved and identified individually (Fig. 2b and c). In previously reported studies using Derenzo phantoms, PET radionuclides such as ⁵²Mn, ¹⁸F, ⁶⁴Cu, ⁴⁸V, ⁴³Sc, ⁸⁹Zr, ⁴⁵Ti, ⁵⁵Co, and ⁶⁸Ga demonstrated excellent relative resolution at 3.2 mm cavities and sufficient visual resolution at 2.4 mm cavities¹⁵. In contrast, ¹¹⁸Te-based PET imaging in the present study did not achieve visual resolution for rods smaller than 2.8 mm (Fig. 2c). Nevertheless, cavities of 3.5 mm and larger were clearly resolved, with a full-width at half-maximum (FWHM) of 2.70 ± 0.06 mm derived from the plot profile of the 3.5 mm cavities, indicating that the spatial resolution obtained with ¹¹⁸Sb is sufficient for evaluating radiotracer distribution in both small-animal and human-scale PET imaging applications.

Long-term PET imaging of ¹¹⁸Sb generated from ¹¹⁸Te

To demonstrate the feasibility of ¹¹⁸Te as an in vivo-PET signal generator, we conducted longitudinal PET imaging of mice. A solution of ¹¹⁸Te was intravenously administered to mice via the tail vein. The mice in each group were maintained for 1, 2, or 3 weeks before undergoing PET/CT scanning.

PET/CT images showed that the radioactivity signals in the abdominal regions were clearly detectable up to 3 weeks (Fig. 3a). After PET imaging, the radioactivity signals in various mouse organs were also measured using a gamma counter. Ex vivo biodistribution analysis demonstrated that the ¹¹⁸Te-derived radioactivity signals in these organs remained detectable even at 3 weeks post-injection. The results confirmed that ¹¹⁸Sb generated from ¹¹⁸Te predominantly accumulated in the liver and spleen (Fig. 3b), which was consistent with the results of PET/CT imaging. Previous studies have suggested that absorbed antimony tends to accumulate in the liver, spleen, and bone¹⁶. Inorganic tellurium compounds are considered to possibly undergo hepatic biotransformation into methylated organotellurium species, which are thought to distribute to the liver, spleen, and kidneys¹⁷. Taken together, these findings suggest that the decay of ¹¹⁸Te into ¹¹⁸Sb may not substantially alter the organ-level localization of radioactivity, indicating a continuity in tissue accumulation across the decay sequence. These data support the capability of ¹¹⁸Te as a tracer for long-term PET imaging.

Fig. 3.

PET/CT images and biodistribution of ¹¹⁸Te in mice over a prolonged period (a) PET/CT fused images at selected slice levels, acquired from ICR female mice at 1, 2, and 3 weeks post injection of ¹¹⁸Te solution. Injected ¹¹⁸Te activities were 0.2 MBq (n = 1), 0.4 MBq (n = 2), and 0.6 MBq (n = 3), respectively. p.i. = post injection (b) After PET imaging, mice were immediately sacrificed and major organs were excised. Radioactivity in each organ was measured using a gamma counter, and the uptake was calculated as %IA/g.

Late-phase time-activity curves (TACs) of ¹¹⁸Te

Late-phase time–activity curves (TACs) derived from ex vivo biodistribution data are presented in Fig. 4. At 1 week p.i., the uptake fraction was highest in the liver (0.105) followed by the spleen (0.0489). The uptake fractions in other tissues and organs were already negligibly small at this time point (≤ 0.003). Both the hepatic and splenic uptake increased thereafter at 2 weeks p.i. and remained the dominant sites of ¹¹⁸Te retention. Although the uptake in the rest remained consistently low at 2 weeks p.i., the heart, lungs, muscle, and bone showed slight increases at this time point. By 3 weeks p.i., hepatic and splenic uptake decreased relative to 2 weeks p.i. but remained higher than those in all other tissues. Uptake in the gastrointestinal tract was generally low, although stomach uptake increased at 3 weeks p.i. with substantial variability. Blood uptake remained negligible at all time points, indicating sustained clearance from circulation.

Fig. 4.

Late-phase time–activity curves (TACs) of ¹¹⁸Te in female Slc: ICR mice derived from ex vivo biodistribution data at 1, 2, and 3 weeks post-injection (p.i.). Sample sizes were n = 1 at 1 week p.i. and n = 2 at 2 and 3 weeks p.i.; data at 2 and 3 weeks p.i. are presented as mean ± standard deviation. The Blood TAC represents uptake fractions normalized to 0.1 g of blood. The corresponding numerical values are provided in Table S1 (Supplementary data).

Discussion

In this study, we used ⁴He²⁺ ion beams to produce the radionuclide ¹¹⁸Te through the nuclear reaction of ¹¹⁶Sn(⁴He,2n)¹¹⁸Te. Historically, ¹¹⁸Te has been recognized as a generator radionuclide for ¹¹⁸Sb. Previous studies have demonstrated in vivo PET imaging using ¹¹⁸Sb extracted from a ¹¹⁸Te/¹¹⁸Sb radionuclide generator (“milking”), the short half-life of ¹¹⁸Sb (T₁/₂ = 3.5 min) permits imaging for a short period only following administration, similar to conventional PET tracers characterized by short half-lives, which inhibit the acquisition of delayed-phase images¹⁸. In contrast, direct use of ¹¹⁸Te provides a steady-state source of ¹¹⁸Sb, enabling sustained in vivo positron signal generation. This strategy supports long-term PET imaging and opens new avenues for longitudinal applications in nuclear medicine. In addition, ¹¹⁸Te doesn’t emit high-energy gamma ray causing decreased resolution of PET image. Despite its favorable nuclear properties, to our knowledge, no prior studies have reported the development of ¹¹⁸Te-labeled compounds or their application in vivo. This lack of precedent is likely attributable to the technical challenges associated with radiolabeling chemistry and the limited availability of ¹¹⁸Te. The current work thus represents a first step toward establishing the feasibility of ¹¹⁸Te-based agents for sustained PET imaging and expanding the toolkit for longitudinal nuclear medicine applications.

Recently, radionuclides with relatively long half-lives, such as ²²⁵Ac (T₁/₂ = 9.9 days) and ¹⁷⁷Lu (T₁/₂ = 6.7 days), have gained considerable attention as therapeutic radionuclides for TRT^19,20. ¹⁷⁷Lu is widely used in clinical field, with approved radiopharmaceuticals such as Pluvicto and Luthathera demonstrating its efficacy in prostate and neuroendocrine tumors, respectively. Inparallel, ¹⁶¹Tb (T₁/₂ = 6.9 days, Eβ⁺_max = 593 keV, Eβ⁺_mean = 154 keV), which has comparable β⁻ energy and half-life to ¹⁷⁷Lu (T₁/₂ = 6.7 days, Eβ⁺_max = 497 keV, Eβ⁺_mean = 134 keV) and additionally emits low-energy electrons such as Auger electrons, has emerged as a promising candidate for TRT. Preclinical studies have shown enhanced therapeutic potential of ¹⁶¹Tb when combined with radioisotope‑drug delivery systems (RI-DDS) targeting CD30 ²¹ or PSMA^22,23. Because the half-lives of ¹¹⁸Te (6.0 days), ¹⁷⁷Lu (T₁/₂ = 6.7 days) and ¹⁶¹Tb (6.9 days) are similar, PET imaging with ¹¹⁸Te-labeled therapeutic agents could enable effective evaluation of the in vivo biodistribution of ¹⁷⁷Lu and ¹⁶¹Tb-labeled therapeutic agents. For example, data obtained by imaging with ¹¹⁸Te‑labeled therapeutic agents can be used to estimate the distribution and accumulation patterns of therapeutic agents labeled with these therapeutic radionuclides during treatment, thereby providing valuable insights for treatment monitoring and the evaluation of side effects arising from nonspecific tissue accumulation.

By comparing with other long half-life positron tracers, ¹¹⁸Te has several benefits. First, ¹¹⁸Te possesses a relatively long half-life (6.0 days), which enables extended imaging windows suitable for tracking carrier molecules with slow pharmacokinetics, such as monoclonal antibodies. This property is comparable to ⁸⁹Zr and ¹²⁴I but may offer additional flexibility depending on the biological context. Published time–activity curves of ⁸⁹Zr‑labeled monoclonal antibodies indicate that reliable tracking is generally limited to within the first week after injection^24,25. However, IgG monoclonal antibodies typically exhibit biological half‑lives of approximately 2–3 weeks²⁶, suggesting that longer‑lived PET radionuclides are required to monitor the full in vivo kinetics of such carrier molecules. Second, unlike ⁸⁹Zr, which emits high-energy gamma photons (e.g., 909 keV) that can contribute to spurious event detection and degrade image quality, ¹¹⁸Te does not emit such photons. This feature may be particularly advantageous for clinical PET imaging, where minimizing background noise.

In this report, PET imaging of positrons emitted from ¹¹⁸Sb generated by ¹¹⁸Te provided relatively diffuse images (Fig. 2) owing to the large average energy of the positrons (1,186 keV). To address this issue, we consider the use of magnetic resonance–positron emission tomography (MR-PET) as a promising solution. The annihilation locations of positrons emitted by radioisotopes are blurred relative to the actual location of the radiation source because of the distance between decay and annihilation. In contrast, when a static magnetic field is applied, positrons undergo spiral trajectories owing to the Lorentz force, which is thought to suppress their lateral spread²⁷. In addition, the compressive effect of the magnetic field on the spatial range of positrons from commonly used positron emitters has been simulated using Geant4²⁸. Furthermore, recent modeling studies have proposed modified semi-empirical formulas to estimate the maximum positron range under varying magnetic field strengths, demonstrating that stronger fields can significantly reduce lateral dispersion of high-energy positrons in PET/MRI systems²⁹. Indeed, simultaneous hybrid MR-PET has been shown to improve the resolution of PET imaging in the plane perpendicular to the static magnetic field of the scanner³⁰. Therefore, it is preferable to use MR-PET to detect this nuclide because the spatial range of the emitted positrons can be suppressed by a magnetic field.

In this study, the observed accumulation of ¹¹⁸Te-derived radioactivity in the liver and spleen may be influenced by several physicochemical factors. ¹¹⁸Te was produced as inorganic tellurium. Under physiological pH conditions such as those in the bloodstream, inorganic tellurium is known to exhibit reduced solubility, which may promote the formation of particulate aggregates. Aggregates within the nanoparticle size range are considered susceptible to phagocytosis by hepatic Kupffer cells and splenic macrophages. This mechanism might explain biodistribution of ¹¹⁸Te, and is consistent with the retention patterns observed in the time–activity curves (TACs). The late-phase TACs showed that the biodistribution of ¹¹⁸Te from 1 to 3 weeks p.i. is dominated by hepatic and splenic retention. The consistently low uptake in kidneys, heart, and muscle indicates minimal redistribution at these time points. The transient increase in skeletal uptake between 1 and 2 weeks p.i., followed by a decrease at 3 weeks p.i., may suggest delayed incorporation in bone, although interpretation is limited by the sparse sampling. Similarly, the increase in stomach uptake at 3 weeks p.i. should be interpreted cautiously and may reflect sample variability rather than systematic accumulation. Although early-phase uptake data are absent, these late-phase TACs provide an insight into ¹¹⁸Te retention patterns at extended time points. Additional measurements at early points are warranted for a more complete characterization of the temporal biodistribution of ¹¹⁸Te.

Despite its promise as a tracer for radiotheranostics, the implementation of ¹¹⁸Te still faces several challenges. This study represents a preliminary assessment of ¹¹⁸Te for in vivo PET imaging, and detailed analyses of its chemical form and stability in solution remain incomplete. To advance its application in targeted molecular systems, it is essential to explore suitable labeling strategies—whether through coordination chemistry or alternative molecular binding approaches—to enable integration of ¹¹⁸Te into targeted drug carriers, such as peptides, antibodies, and nanocarriers. Previous analytical studies have demonstrated that Tellurium can be chelated by sulfur-containing compounds such as diethyldithiocarbamates and dithizone³¹. This finding suggests that sulfur containing compounds may offer a promising scaffold for developing ¹¹⁸Te-labeled carrier agents. Also, given the relatively long physical half-life of ¹¹⁸Te (6.0 days), prolonged radiation exposure following systemic administration may raise safety concerns. Therefore, future studies must include dosimetry evaluation and radiotoxicity assessment, particularly regarding nonspecific tissue accumulation and clearance kinetics. These investigations will be critical for optimizing dosing strategies and ensuring safe translation to clinical applications. In parallel, comparative evaluation with existing long-lived positron emitters, such as ⁸⁹Zr, is warranted to determine the advantages offered by ¹¹⁸Te in terms of radiolabeling versatility, imaging timeframe, and compatibility with biomolecular platforms. The production of ¹¹⁸Te requires a relatively high-energy cyclotron capable of accelerating ⁴He ions to approximately 30 MeV, such as an AVF cyclotron. In contrast, ⁸⁹Zr is routinely produced via proton irradiation of ⁸⁹Y targets using standard medical cyclotrons in clinical PET facilities.

With the recent expansion of research into radiotheranostics and its clinical market³², therapeutic strategies increasingly rely on radionuclides and carrier molecules with prolonged biological and physical half-lives. Advancement of radiotheranostics will require long-lived diagnostic radionuclides to complement therapeutic counterparts and enable synchronized treatment monitoring.

Materials and methods

Production of ¹¹⁸Te

¹¹⁸Te was produced through the nuclear reaction of ¹¹⁶Sn(⁴He,2n)¹¹⁸Te. An enriched Sn-116 metallic foil with a thickness of 21 μm was bombarded for 3.9 h with a 27.9 MeV ⁴He²⁺ beam delivered by the AVF cyclotron at RIKEN RI Beam Factory, Wako, Japan. The isotopic composition of the enriched ¹¹⁶Sn target was < 0.06% ¹¹²Sn, < 0.03% ¹¹⁴Sn, 0.06% ¹¹⁵Sn, 97.8% ¹¹⁶Sn, 0.90% ¹¹⁷Sn, 0.67% ¹¹⁸Sn, 0.11% ¹¹⁹Sn, 0.41% ¹²⁰Sn, 0.03% ¹²²Sn, and 0.02% ¹²⁴Sn. The current of the ⁴He²⁺ beam was 9.8 electric µA. During irradiation, the target was directly cooled with helium gas and water and indirectly cooled through a 1 mm-aluminum backing plate from the upstream and downstream side of the beam, respectively.

After irradiation, ¹¹⁸Te was chemically separated from the target material by coprecipitation with metallic palladium, followed by anion-exchange column chromatography, according to a procedure reported previously³³. Our experience in ^121mTe separation³⁴ led us to apply Muromac 1 × 8 (200–400 mesh) in this work and Dowex 2 × 8 (200–400 mesh) in the previous study³³ for anion-exchange chromatography. Finally, a solution (290 µL) of no-carrier-added ¹¹⁸Te in 0.01 M HCl, including 18 mg of sodium chloride, was prepared. The γ-ray spectrum of purified ¹¹⁸Te (Figure S1) was acquired with a germanium semiconductor detector 7.7 days after the end of the bombardment. The γ-peaks attributed to ¹¹⁸Te, daughter nuclide ¹¹⁸Sb, and ^119mTe, a byproduct of the ¹¹⁶Sn(⁴He, n)¹¹⁹Te reaction, are indicated in Fig. S1. The radioactivities of ¹¹⁸Te and ^119mTe were 4.5 and 3.9 MBq, respectively, and the radiochemical purity of ¹¹⁸Te was calculated as 54% at this time point. According to evaluated nuclear data³⁶, ^119mTe decays predominantly by electron capture with no meaningful positron emission and therefore does not contribute to PET signals. The specific radioactivity of ¹¹⁸Te was evaluated as > 10 MBq/µg using an ICP-MS (Inductively Coupled Plasma Mass Spectrometer, Agilent 7700) to determine Te concentration. ¹¹⁸Te decays to ¹¹⁸Sb with a half-life (T_1/2) of 6.02 days by electron capture (EC), and ¹¹⁸Sb immediately decays to stable ¹¹⁸Sn with T_1/2 = 3.61 min by β⁺ decay (decay branch b_β = 73.5%) and EC decay (b_EC = 26.5%) (Fig. 1). The γ rays (511.0 keV) of positron annihilation following the β⁺ decay of ¹¹⁸Sb in radioactive equilibrium with ¹¹⁸Te are useful for PET imaging of ¹¹⁸Te. Purified ¹¹⁸Te was shipped to the Isotope Science Center at the University of Tokyo for animal experiments.

PET imaging analysis of phantom

NEMA and Derenzo phantoms were filled with 1.0 MBq ¹¹⁸Te in 0.1 M HCl. PET images were acquired for 3 h using a small-animal PET scanner (Clairvivo, Shimadzu, Japan). Acquired raw data were reconstructed using the List-DRAMA algorithm and finally exported as DICOM files. Those files were analyzed and converted to image files using a DICOM viewer (OsiriX V14.1 MD, Pixmeo, Switzerland). For detailed assessment of the spatial resolution, ImageJ (www.imagej.net) was used to extract line profiles across the 4.3 mm and 3.5 mm cavities of the Derenzo phantom (Fig. 2B). Furthermore, full-width at half-maximum (FWHM) values were calculated from plot profiling data of three distinct signal peaks within the 3.5 mm cavities.

In vivo PET imaging

All animal experiments were approved by the Institutional Review Board of Animal Research at the Isotope Science Center, The University of Tokyo, and were conducted in accordance with institutional (University of Tokyo Conduct no. 129: Code of Practice for Animal Research) and ARRIVE’s guidelines³⁵. Anesthesia and euthanasia procedures were conducted in accordance with the American Veterinary Medical Association (AVMA) Guidelines for the Euthanasia of Animals (2020). All methods of animal experiments in this study were performed in accordance with the relevant guidelines and regulations.

Female Slc: ICR outbred mice (8 weeks old, weighing 27–32 g) were obtained from Japan SLC Inc. (Japan). All mice used in this experiment were housed under a cycle of 12 h of light and 12 h of dark with ad libitum access to food and water. Environmental conditions were maintained at 20–25 °C and 50–60% relative humidity to ensure physiological stability.

A solution of ¹¹⁸Te diluted with 0.05 M HCl was adjusted to a final injection volume of 150 µL per mouse. To assess PET imaging at different post-injection intervals, ¹¹⁸Te solutions were administered via the tail vein into the mice at doses of 0.2 MBq (n = 1), 0.4 MBq (n = 2), and 0.6 MBq (n = 2). Subsequently, PET scans were performed at 1 week, 2 weeks, and 3 weeks after injection. Isoflurane anesthesia was administered using a vaporizer system (SN-487, Shinano Seisakusho Co., Japan) for both induction and maintenance. For induction, mice were placed in a small plastic chamber connected to SN-487 and exposed to 3–4% isoflurane in oxygen. After induction, each mouse was transferred to the PET/CT imaging stage and fitted with an anesthesia mask connected to the same vaporizer system while mouse body temperature was maintained at 37 °C using a temperature-controlled heating system. Anesthesia was maintained at 1.5–2% isoflurane throughout the PET and CT imaging procedures. Images of mice under isoflurane anesthesia were acquired using a Clairvivo PET scanner (Shimadzu) for a duration of 2 h. Immediately following PET imaging, anatomical imaging of mice maintained under anesthesia was performed using a Clairvivo CT scanner (Shimadzu). Raw PET data were reconstructed using the List-DRAMA algorithm implemented in the Clairvivo PET system (Shimadzu). Reconstructed images were exported as DICOM files. Moreover, raw CT data were converted into the DICOM format using the Clairvivo CT system (Shimadzu). Subsequently, PET and CT DICOM files were co-registered and fused using the OsiriX DICOM viewer (Pixmeo, Switzerland) to generate integrated PET/CT images.

Ex vivo biodistribution analysis

Following PET scanning, mice were euthanized under deep anesthesia induced by 3–4% isoflurane using the vaporizer system (AnesⅡ, Biomachinery Co., Japan). Under anesthesia, cardiac puncture was performed to collect blood samples. Immediately afterward, cervical dislocation was carried out to ensure complete euthanasia by trained and experienced researchers, in accordance with The University of Tokyo Manual for Animal Experiments. Blood, tissues, and organs (liver, spleen, kidney, stomach, large intestine, small intestine, heart, lung, muscle, and bone) were collected and weighed. The radioactivity of these tissues and organs was measured using a gamma counter (Cobra Quantum, PerkinElmer, USA). Radioactive counts were decay-corrected to calculate the percentage of the injected activity per gram of tissue (%IA/g).

Time-activity curves (TACs) analysis

Late-phase TACs were constructed using uptake fractions dervied from the ex vivo biodistribution data at 1, 2, and 3 weeks p.i. Blood uptake was normalized to 0.1 g to account for variability in the weight of collected blood samples (0.535–0.946 g). These TACs were used to explore the late-phase retention trends of ¹¹⁸Te.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

This study was conceived and designed by HT. Experiments were carried out by SM, HH, MK, YK, AO-N, AH, XQ, and NA. Data were analyzed by SM, TM, IOU, NA, YK, HH, HT, and JJ. This manuscript was written by SM, HH, TM, MK, IOU, NA, JJ, and YW.

Funding

This research was financed by the commissioned research fund provided by F-REI (JPFR2504030). This work was partly supported by JSPS KAKENHI Grant Number 22H04961, 22KK0108 and 25K18923.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.