Visual Abstract
Abstract
The aim of this work was to explore 132La as a PET imaging surrogate for 225Ac using a DOTA-based, tumor-targeting alkylphosphocholine (NM600). Methods: 132La was produced on a biomedical cyclotron. For in vivo experiments, mice bearing 4T1 tumors were administered 132La-NM600, and PET/CT scans were acquired up to 24 h after injection. After the last time point, the ex vivo tissue distribution was measured to corroborate the in vivo PET data. The ex vivo tissue distribution in mice was determined at 4 and 24 h after injection of 225Ac-NM600. Results: PET/CT images showed elevated, persistent 132La-NM600 uptake in the tumor. Low bone accumulation confirmed the in vivo stability of the conjugate. Ex vivo biodistribution studies validated the image-derived quantitative data, and the comparison of the 132La-NM600 and 225Ac-NM600 tissue distributions revealed a similar biodistribution for the 2 radiotracers. Conclusion: These findings suggest that 132La is a suitable imaging surrogate to probe the in vivo biodistribution of 225Ac radiotherapeutics.
Targeted radionuclide therapy is a powerful systemic approach to cancer treatment. Recent clinical evidence demonstrates its potential to revolutionize the management of advanced-stage cancers (1,2). To date, 225Ac (half-life, 9.92 d; 100% α-emission; α-energy, 5.78 MeV) is one of the most promising radionuclides for targeted radionuclide therapy (3). Its physical and chemical properties are suitable for the generation of radiopharmaceuticals based on small molecules, peptides, antibodies, and antibody fragments. Logistically, the long half-life of 225Ac enables centralized production and global distribution. However, the lack of isotopes of actinium with imageable emissions makes the noninvasive characterization of 225Ac pharmaceuticals’ biodistribution difficult and requires surrogate radiometals for imaging (2,4). Imaging of 213Bi (half-life, 45.62 min; 97.80% β−-emission; 2.20% α-emission; 25.94% 440.45-keV γ-emission), 225Ac’s third radioactive daughter, using SPECT has been proposed (5), but quantitation is difficult given the lack of radioactive equilibrium in vivo. Other radionuclides (e.g., 68Ga) with significant chemical dissimilarity to actinium are currently used for this purpose, despite reports that the choice of radionuclide can affect the pharmacokinetics of radiopharmaceuticals (6). We and others hypothesize that lanthanum-labeled agents’ biodistribution mirrors actinium agents’ biodistribution in vivo (7–9), but this hypothesis is yet to be tested. Our goal is to evaluate the potential of positron-emitting 132La (half-life, 4.59 h, 42% β+-emission) as a PET imaging surrogate for 225Ac with a well-characterized, tumor-targeting alkylphosphocholine (NM600) developed by our group and used for targeted radionuclide therapy applications with 90Y, 177Lu, and 225Ac (10–12). The compound has selective tumor uptake and retention in a wide variety of cancer types, including the murine mammary adenocarcinoma 4T1 model used in this work.
MATERIALS AND METHODS
All solutions were prepared using 18 MΩ cm−1 water and Optima-grade HNO3 and HCl from Fisher Chemical. The barium target material was acquired from Sigma-Aldrich (99.9%, trace metal basis) and stored in an argon atmosphere. The synthesis of 2-(trimethylammonio)ethyl(18-(4-(2-(4,7,10-tris(carboxymethyl)-1,4,7,10-tetraazacyclododecan-1-yl)acetamido)phenyl)octadecyl) phosphate (NM600) has been described elsewhere (11).
Radiochemistry and Stability
132La was produced by proton irradiation of metallic natBa targets via the natBa(p,n)132La reaction on a 16-MeV GE Healthcare PETtrace cyclotron (7). Irradiations of natBa also produce the low-energy electron-emitting radionuclide 135La (half-life, 19.93 h, 100% electron capture). Under these irradiation conditions (11.9 MeV, 20 μA, 3 h), typical 132La and 135La decay-corrected physical yields were 0.26 and 5.6 MBq μA−1, respectively (7,8). Radiochemical isolation of 132/135La3+ (13xLa3+) was performed by precipitation and single-column extraction chromatography as previously described (7). 225Ac was obtained as 225Ac(NO3)3 from Oak Ridge National Laboratory.
For radiolabeling of NM600 with 13xLa or 225Ac, 340–370 MBq of 13xLa and 3.7 MBq of 225Ac were mixed with NM600 (3.7 MBq/nmol) in 0.1 M NaOAc buffer solution (pH 5) at 80°C for 30 min under constant shaking (500 rpm). Complexes were purified by solid-phase extraction chromatography using Sep Pak C18 cartridges (Waters), eluted in absolute ethanol, dried, and reformulated in phosphate-buffered saline for injection. 13xLa or 225Ac activities were quantified by high-purity germanium spectrometry (aluminum-windowed; full width at half maximum, 1.8 keV at 1,333 keV). The radiochemical yield was assessed by thin-layer chromatography using 50 mM ethylenediaminetetraacetic acid as the mobile phase. To assess stability, 13xLa-NM600 and 225Ac-NM600 complexes were incubated at 37°C in whole mouse serum, and samples were analyzed by thin-layer chromatography after 4 and 24 h of incubation (n = 3). 225Ac-NM600 thin-layer chromatography samples were counted 4 h after spotting of the plates to ensure secular equilibrium conditions. The radiochemical purity and stability of the 13xLa-NM600 complex were confirmed via radio–high-performance liquid chromatography using a reverse-phase 250 × 3.00 mm C18 Luna 5-μm, 100-Å column (Phenomenex) and a water:acetonitrile gradient (5% MeCN, 0–2 min; 5%–65% MeCN, 2–30 min; 65%–90% MeCN, 30–35 min; 90%–5% MeCN, 35–45 min).
PET/CT Imaging and Ex Vivo Biodistribution
Murine mammary adenocarcinoma 4T1 cell lines were obtained from ATCC. The cells were cultured in completed medium (RPMI-1640) supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin in an incubator at 37°C and a 5% CO2 atmosphere. All animal experiments were performed under the approval of the University of Wisconsin Institutional Animal Care and Use Committee. Tumors were subcutaneously grafted by injecting 5.0–7.5 × 105 4T1 cells into the right lower flank of 6- to 8-wk-old female BALB/c mice (Envigo). The mice were used for in vivo imaging and ex vivo biodistribution studies approximately 2 wk after implantation, when the tumor volume reached 400–600 mm3.
For in vivo PET/CT imaging, the mice were intravenously administered 3.7 MBq of 132La-NM600 (30 nmol) (n = 3) in 200 μL of phosphate-buffered saline. After isoflurane anesthesia (2%–4%), PET scans with 40–80 million coincidence events (time window, 3.432 ns; energy window, 350–650 keV) were acquired at 4, 10, and 24 h after injection of the radiotracer. CT scans were performed before each PET acquisition for attenuation correction and anatomic coregistration. Quantitative region-of-interest analysis of the PET images was performed by manually drawing regions of interest to quantify the radionuclide uptake in the tumor and in healthy organs and tissues. After the last imaging time point at 24 h after injection, the ex vivo tissue distribution was measured to corroborate the in vivo PET data and to further examine the similarity between 132La-NM600 and 225Ac-NM600. The ex vivo tissue distribution was also determined in 3 mice at 4 and 24 h after intravenous injection of 225Ac-NM600, 0.01 MBq (0.003 nmol) in 200 μL of phosphate-buffered saline. The mice were sacrificed by CO2 asphyxiation, and tumors and other tissues were collected, wet-weighed, and counted in a calibrated automatic γ-counter (Wizard 2; Perkin Elmer). 13xLa was quantified using an energy window from 10 to 100 keV. 225Ac samples were counted 4 h after organ harvesting to ensure secular equilibrium conditions using an energy window from 50 to 550 keV. Quantitative data are reported as percentage injected activity per gram of tissue (%IA/g) (mean value ± SD).
RESULTS
Radiochemistry and Stability
Both 13xLa-NM600 and 225Ac-NM600 were produced in almost quantitative radiochemical yields (>95%) at a similar DOTA-based apparent molar activity of 3.7 MBq/nmol. The stability of both radioconjugates was evaluated by radio–thin-layer chromatography for up to 24 h after incubation. Stabilities higher than 98% were observed, indicating minimal degradation of 13xLa-NM600 and 225Ac-NM600 in whole mouse serum (Fig. 1A; Supplemental Table 1 [supplemental materials are available at http://jnm.snmjournals.org]).
(A) Maximum-intensity-projection static PET images of representative BALB/c mouse injected intravenously with 132La-NM600. (B) 132La-NM600 tissue uptake quantification of PET regions of interest in mice (n = 3, mean ± SD) injected with 132La-NM600. Ex vivo 225Ac-NM600 biodistribution data at 4 h are also included for comparison.
The radiochemical purity and serum stability of 13xLa-NM600 were corroborated by radio–high-performance liquid chromatography (Fig. 1B). A radiopeak (retention time, 26.04 min) corresponding to the 13xLa-NM600 complex was observed in all radiochromatograms. Moreover, no additional radiopeaks were noted either in the solvent front (retention time, <3 min) or at different retention times, indicating the absence of free 13xLa3+ or 13xLa-NM600 degradation products. Similar results were observed after incubation of 13xLa-NM600 in mouse serum for 24 h.
In Vivo Imaging and Ex Vivo Biodistribution
Figure 2A shows representative maximum-intensity-projection PET images after intravenous injection of 132La-NM600 in BALB/c mice bearing 4T1 tumors. 132La-NM600 uptake in the tumor was elevated and persistent, reaching 11.5 ± 1.3 %IA/g at 24 h after injection (Fig. 2B; Supplemental Table 2). Blood-pool activities were initially high (17.8 ± 1.3 %IA/g) and gradually declined to 13.8 ± 1.3 and 7.3 ± 0.8 %IA/g at the two later time points. Accumulation in the liver peaked at 18.3 ± 0.5 %IA/g 10 h after injection, and uptake in the kidneys was relatively low (<7 %IA/g) at all imaging time points. Accumulation in other normal tissues and organs remained below 5 %IA/g. These imaging results agreed with the 225Ac-NM600 distributions obtained by ex vivo γ-counting (Fig. 2B; Supplemental Table 3). Similarly, elevated blood uptake (16.8 ± 0.8 %IA/g) and liver uptake (11.9 ± 0.6 %IA/g) were seen for 225Ac-NM600 at 4 h after injection. Blood values declined over the next 24 h to 5.7 ± 0.3 %IA/g, and liver uptake increased to 17.5 ± 0.3 %IA/g. An increase in 225Ac-NM600 uptake in the tumor from 4.6 ± 0.4 %IA/g to 11.5 ± 2.1 %IA/g was also observed during the same period. Ex vivo biodistribution studies after the last PET/CT scan (Fig. 3) confirmed the image-derived quantitative data showing elevated 132La-NM600 uptake in the tumor (11.1 ± 0.8 %IA/g), liver (17.0 ± 0.9 %IA/g), and blood (9.9 ± 0.2 %IA/g). A more extensive comparison of 132La-NM600 and 225Ac-NM600 tissue distribution at 24 h after injection revealed the 2 radiotracers to have an effectively identical biodistribution.
(A) 13xLa-NM600 and 225Ac-NM600 mouse serum stability quantified by radio–thin-layer chromatography (n = 3, mean ± SD) (B) Radio–high-performance liquid chromatography chromatogram of 13xLa-NM600 radiolabeling reaction (3.7 MBq/nmol) after incubation time of 30 min (top) and 13xLa-NM600 serum stability chromatogram at 24 h after incubation (bottom).
Ex vivo 132La-NM600 and 225Ac-NM600 biodistribution at 24 h after injection, measured by γ-counting (n = 3, mean ± SD).
DISCUSSION
Lanthanides, and specifically La3+, have been identified as suitable nonradioactive surrogates for Ac3+ (13). Chemical similarities between these 2 ions have been exploited to unravel the chemistry of actinium, and stable lanthanum has been instrumental to the design of novel chelators that stably retain 225Ac in vivo (13). Although potential for using different radiolanthanums as PET imaging surrogates for 225Ac has been proposed (7–9), to date, no direct comparison between the in vivo distribution of a radioactive compound labeled with actinium or lanthanum has been performed. Thanks to a half-life of several hours, a higher positron branching ratio, and simpler, scalable production on low-energy cyclotrons, we chose 132La from the unstable lanthanum radioisotopes with positron emissions. Coproduced 135La can be minimized with irradiation of isotopically enriched 132Ba; however, the results of the proof-of-concept imaging studies reported herein are not affected by the presence of 135La.
Our observed in vitro stability of 13xLa-NM600 and 225Ac-NM600 agrees with previously reported stabilities for DOTA compounds with 225Ac and 140La, showing that both DOTA complexes were stable for several days after incubation in human or complete mouse serum (12,14). High tumor uptake in vivo and ex vivo confirms the tumor selectivity of the targeting vector. Hepatobiliary excretion of the tracer is also apparent. The observed low bone accumulation indicates the in vivo stability of the 132La-NM600 radioconjugate: free radiolanthanum has been shown to lodge in the bones in mice (7). Ex vivo biodistribution analysis further validates the image-derived quantitative data, and the similar 132La-NM600 and 225Ac-NM600 tissue distributions support the hypothesis that 132La is a suitable imaging surrogate for 225Ac-targeted radiotherapeutic drugs in vivo.
One limitation of 132La is its 4.59-h half-life, compared with the 9.92-d half-life of 225Ac. This difference is not critical for fast-clearing small molecules such as prostate-specific membrane antigen, fibroblast activation protein inhibitor, or DOTATATE, among many other clinically relevant small biomolecules. However, in the case of macromolecules with longer biologic half-lives, such as antibodies and antibody derivatives, 132La will likely not be suitable for protracted imaging time points. Nonetheless, evidence suggests that the effects of the radioisotope’s chemical properties on the pharmacokinetic profile of these compounds is less pronounced, in which case other trivalent imaging surrogates, such as 86Y3+ and 111In3+, may be appropriate (11,15).
CONCLUSION
This work demonstrates the potential of 132La as a PET imaging surrogate to probe the in vivo biodistribution of 225Ac radiotherapeutics. Our results warrant future development of the theranostic 132La/225Ac pair, which will require isotopically enriched 132Ba targets to produce larger quantities of highly pure 132La.
DISCLOSURE
Eduardo Aluicio-Sarduy receives financial support from the National Institutes of Health under award T32CA009206. Jamey Weichert is the founder and CSO of Archeus Technologies Inc., which has licensing rights to NM600. Jamey Weichert and Reinier Hernandez have an equity stake in Archeus Technologies Inc. No other potential conflict of interest relevant to this article was reported.
KEY POINTS
QUESTION: Is the positron-emitter 132La a suitable PET imaging surrogate to probe the in vivo biodistribution of 225Ac radiotherapeutics?
PERTINENT FINDINGS: 132La PET image–derived and ex vivo pharmacokinetics match the measured biodistribution of 225Ac in a murine model of breast cancer targeted with a DOTA-based small molecule (NM600) throughout the usual lifetime of 132La.
IMPLICATIONS FOR PATIENT CARE: These findings and the straightforward cyclotron production of positron-emitting 132La make it an attractive candidate for rapid, noninvasive screening of 225Ac radiopharmaceuticals.
ACKNOWLEDGMENTS
We thank Archeus Technologies Inc. for providing the NM600 precursor, and we thank the staff of the University of Wisconsin Small Animal Imaging Facility and the University of Wisconsin Carbone Cancer Center for their support.
Footnotes
↵* Contributed equally to this work.
Published online Oct. 30, 2020.
- © 2021 by the Society of Nuclear Medicine and Molecular Imaging.
REFERENCES
- Received for publication August 24, 2020.
- Revision received October 14, 2020.
