Abstract
Expression of the chemokine receptor chemokine C-X-C motif receptor 4 (CXCR4) plays an important role in cancer metastasis, in autoimmune diseases, and during stem cell–based repair processes after stroke and myocardial infarction. Previously reported PET imaging agents targeting CXCR4 suffer from either high nonspecific uptake or bind only to the human form of the receptor. The objective of this study was to develop a high-stability 64Cu-labeled small-molecule PET agent for imaging both human and murine CXCR4 chemokine receptors. Methods: Synthesis, radiochemistry, stability and radioligand binding assays were performed for the novel tracer 64Cu-CuCB-bicyclam. In vivo dynamic PET studies were performed on mice bearing U87 (CXCR4 low-expressing) and U87.CXCR4 (human-CXCR4 high-expressing) tumors. Biodistribution and receptor blocking studies were performed on CD1-IGS immunocompetent mice. CXCR4 expression on tumor and liver disaggregates was confirmed using a combination of immunohistochemistry, quantitative polymerase chain reaction, and Western blot. Results: 64Cu-CuCB-bicyclam has a high affinity for both the human and the murine variants of the CXCR4 receptor (half-maximal inhibitory concentration, 8 nM [human]/2 nM [murine]) and can be obtained from the parent chelator that has low affinity. In vitro and in vivo studies demonstrate specific uptake in CXCR4-expressing cells that can be blocked by more than 90% using a higher-affinity antagonist, with limited uptake in non–CXCR4-expressing organs and high in vivo stability. The tracer was also able to selectively displace the CXCR4 antagonists AMD3100 and AMD3465 from the liver. Conclusion: The tetraazamacrocyclic small molecule 64Cu-CuCB-bicyclam has been shown to be an imaging agent for the CXCR4 receptor that is likely to be applicable across a range of species. It has high affinity and stability and is suitable for preclinical research in immunocompetent murine models.
Chemokine C-X-C motif receptor 4 (CXCR4) is a 7-transmembrane helix G-protein–coupled receptor. The interaction of CXCR4 with its cognate ligand, stromal cell–derived factor 1α (SDF-1α/CXCL12), is essential during embryonic development and plays a key role in normal physiologic function, including control of hematopoietic cells during homeostasis (1). CXCR4 is overexpressed in multiple cancer types, with expression levels associated with increased proliferation, migration, and survival (2,3). In addition, CXCR4 expression on circulating tumor cells has been shown to enable metastasis to CXCL12-expressing organs, including liver, lung, and bone marrow (4), and inhibition of CXCR4/CXCL12 signaling reduces metastasis in various breast cancer models (5). There is a clear role in multiple myeloma, particularly in extramedullary disease, a chemotherapy-resistant subtype with poor prognosis (6,7). Important links have been elucidated between CXCL12 and immunotherapies, where blocking the signaling axis prevented recruitment of immunosuppressing fibroblast activation protein–positive stromal cells (8). CXCR4 has also been shown to have a role in stroke, autoimmune disease (9,10), and myocardial infarction, with CXCL12 upregulated after infarction (11–13). CXCR4 is therefore a candidate prognostic biomarker for several clinical indications. Because several CXCR4-targeted agents are currently in clinical trials, CXCR4 also has potential as a predictive biomarker to enable patient stratification.
The 68Ga-radiolabeled cyclic peptide CPCR4.2 (14,15), pentixafor, has been used to demonstrate the potential of CXCR4 PET imaging over the past few years. Pentixafor has been used clinically to image human CXCR4 expression in a range of cancer types, including lymphoma, multiple melanoma, glioblastoma, and small cell lung cancer in proof-of-concept studies (16–21). However, pentixafor has little or no binding affinity to the murine variant of CXCR4 (18), limiting assessment of nonxenogenic CXCR4 expression in preclinical models.
Small-molecule CXCR4 antagonists based on tetraazamacrocycles, including AMD3100 and AMD3465 (22) (Supplemental Fig. 1; supplemental materials are available at http://jnm.snmjournals.org), bind to both the human and the mouse variants of the receptor (23–26), with AMD3100 the sole CXCR4 antagonist currently approved by the U.S. Food and Drug Administration and the European Medicines Agency for human use. The cyclam components of these molecules allow for transition metal complex formation, which switches the binding mode from electrostatic/H-bonding between the protonated cyclam amine groups and aspartate residue side chains (171 and 262) to coordinate bonds with the same residues. Direct labeling with 64Cu to enable PET imaging has been attempted (27–30), and although CXCR4-dependent tumor uptake of 64Cu-CuAMD3100 and 64Cu-CuAMD3465 (Supplemental Fig. 1) has been demonstrated, high nonspecific hepatic accumulation (likely caused by complex instability) prevented further development of these compounds (28,29). Copper(II) cyclam complexes are unlikely to have sufficient kinetic stability to retain the metal ion in vivo (31–38).
Configurational restriction of the cyclam with an ethylene cross-bridge confers kinetic stability on the copper(II) complex and increases affinity for, and residence time at, the CXCR4 receptor. The ethylene bridge in the macrocyclic rings of CB-bicyclam (Fig. 1B) allows only 1 configuration for the macrocycle copper(II) complex, which optimizes coordinate bond formation with CXCR4 residues relative to nonbridged cyclam metal complexes of AMD3100 (37,39–44). The unbridged macrocycle complexes are in equilibrium between up to 6 possible configurations, leading to a stochastic overall affinity reflective of the average of these states. The affinity increase is likely mediated by reduction in bond lengths (effectively switched from axial to equatorial) to give shorter, stronger coordination bonds to CXCR4 receptor surface aspartate residues (39,40).
In vitro assessment of 64Cu-CuCB-bicyclam. (A) Comparative affinity values vs. CXCL12 determined by calcium flux assay. (B) Chemical structures of 64Cu-CuAMD3100 and 64Cu-CuCB-bicyclam. (C) Fluorescence-activated cell sorting determination of CXCR4 expression. (D) Binding of 64Cu-CuCB-bicyclam on U87 and U87.CXCR4-expressing cell lines.
We have developed an understanding of structure–activity relationships of CXCR4-binding bridged azamacrocyclic compounds and set out to apply this knowledge to the development of PET imaging probes. Currently, there are no blockable and stable nuclear imaging agents that bind to both human and nonhuman variants of CXCR4, inhibiting research in syngeneic models or understanding of the role of CXCR4 in developmental pathways (45). In this work, we investigated a 64Cu-labeled tetraazamacrocyclic small-molecule CXCR4 antagonist that has suitable characteristics (biodistribution, specificity, and stability) for imaging both the human and the murine homologs of the CXCR4 receptor.
MATERIALS AND METHODS
Ethics
All animal procedures were approved by the University of Hull Animal Welfare Ethical Review Body and performed in accordance with the United Kingdom’s Guidance on the Operation of Animals (Scientific Procedures) Act of 1986 and within guidelines set out by the United Kingdom National Cancer Research Institute Committee on the Welfare of Animals in Cancer Research under Home Office Project License number 60/4549 held by Dr. Cawthorne (46).
General
All chemicals were supplied by Sigma-Aldrich unless otherwise stated and were used as supplied. Elemental analysis was performed using a carbon, hydrogen, and nitrogen analyzer (EA1108; Carlo Erba). Accurate mass spectrometry measurements were obtained using a linear-trap-quadropole device (Orbitrap XL; Thermo Fisher Scientific). Semipreparative high-performance liquid chromatography was performed using an Agilent 1100 series device equipped with an ultraviolet light detector (series G1314A) and a NaI radiodetector. Thin-layer chromatography was performed using neutral alumina sheets (Merck), eluting with 95:5 methanol:water, saturated with excess NaCl. Radio–thin-layer chromatography was performed using a Lablogic ScanRam, equipped with a NaI detector at a speed of 10 mm/min. Data were recorded using Lablogic Laura (version 4.1.13.91). 64Cu was purchased from the University of Cambridge. Further details of synthesis, radiochemistry, and stability methodologies are presented in Supplemental Figures 2 and 3. 64Cu-CuAMD3100 was synthesized and analyzed following a reported procedure (27).
Cell Lines
The U87 cell line was purchased from American Type Culture Collection, and the glioblastoma astrocytoma U87.CD4 cell line was transfected with human CXCR4 or murine CXCR4. Both U87 and U87.CXCR4 (human and murine) cells were grown in Dulbecco modified Eagle medium, with l-glutamine medium (Lonza) with U87.CXCR4 supplemented by a 2 μg/mL concentration of puromycin. All cell lines were cultured in the presence of 10% fetal bovine serum, in a humidified incubator with 5% CO2. The intracellular calcium mobilization in response to CXCL12 was measured in human U87.CXCR4 cell lines at 37°C by monitoring the fluorescence of an intracellular calcium probe (Fig. 1A) as a function of time using a fluorometric imaging plate reader (Molecular Devices) (40,47), with compounds (AMD3100, CuAMD3100, CB-bicyclam, CuCB-bicyclam, and Cu2CB-bicyclam) tested in the concentration range 0.01–10,000 nM. Flow cytometry was performed as previously described using a FACSCalibur (Becton Dickinson), and data were analyzed using CellQuest software (Becton Dickinson) (39). CXCR4 expression was determined by Western blot, immunohistochemistry (as previously described) (15,28), and quantitative polymerase chain reaction (as previously described; Supplemental Table 1).
Receptor Binding Assay
Human U87.CXCR4 cells grown to 60%–80% confluence were used for receptor binding assays. Cells were washed with cold phosphate-buffered saline (PBS), and flasks were placed on ice to prevent receptor internalization. Cells were detached using a cell scraper and centrifuged at 200g for 5 min; 5 × 105 cells were then resuspended with cold PBS. Cells were incubated with 64Cu-CuCB-bicyclam (37 kBq/mL; 5.2 nM) in PBS for 60 min at 4οC. After incubation, cells were washed quickly 3 times with cold PBS and cell-associated radioactivity was determined in using an automatic γ-counter (Wizard 3″; Wallac).
For blocking experiments, cells were preincubated with 20 μM Cu2CB-bicyclam for 1 h before being washed 3 times in cold PBS buffer and incubated with radiotracer as above. Data are expressed as percentage incubated dose ± SEM, representing the mean of 3 independent experiments with 3 internal repeats.
Animal Models
Female CD1-IGS (Crl:CD1(ICR)) and CD1-Nude (Crl:CD1-Foxn1Nu) mice (aged 21–27 d; weight, 20–25 g) were purchased from Charles River Laboratories. CD1-Nude mice were subcutaneously implanted with a 5 × 106 cell/100 μL suspension of U87/U87-human CXCR4 cells in Geltrex (Life Technologies) basement membrane in the upper flank, under anesthesia. Tumor sizes were measured every 2 d using calipers, and tumor volume (mm3) was determined as length × width × height.
PET/CT Imaging and Analysis
Dynamic whole-body PET and CT images were acquired on a SuperArgus 2R PET scanner (Sedecal). Anesthesia was induced with 5% isoflurane/oxygen (v/v) and then maintained at 2%, using a flow rate of 1 L/min. After cannulation of the tail vein using a bespoke catheter, the mice were placed into an imaging cell where temperature and respiration were monitored (Minerve). 64Cu-CuCB-bicyclam (8.9 ± 3.2 MBq, 7.15 GBq/μmol, 1.2 nmol) was injected at the beginning of a 90-min dynamic imaging sequence (10 frames: 3 × 120 s, 1 × 240 s, 4 × 600 s, and 2 × 1,200 s). The mice were kept at 1% anesthesia during scanning, with temperature and respiration monitored throughout. After the PET scan, a CT image was acquired for anatomic coregistration (40 kV, 140 μA, 360 projections, 8 shots).
For specificity studies, a separate group of mice was injected with a 5 mg/kg blocking dose of Cu2CB-bicyclam administered intraperitoneally 60 min before injection of 64Cu-CuCB-bicyclam.
PET emission data were reconstructed using 3-dimensional ordered-subset expectation maximization with corrections for randoms, scatter. and attenuation. Data were analyzed using AMIDE (48) and Vivoquant software (InVicro), with regions of interest drawn over tumors and various tissues to generate time–activity curves. SUVs were obtained after correction for injected dose and animal weight. Tumor and liver samples were taken at sacrifice and fixed in formalin or frozen in liquid nitrogen.
Ex Vivo Biodistribution, Urine Stability, and Blocking Studies
CD1-IGS mice were injected intraperitoneally with a 5 mg/kg dose of Cu2CB-bicyclam, CuCB-bicyclam, AMD3100, or AMD3465 in saline. For control experiments, no blocking injection was given. At 60 min after blocking, the mice were injected intravenously with approximately 740 kBq of 64Cu-CuCB-bicyclam (or 64Cu-AMD3100) in the tail vein, and this stock solution was used to calibrate the automatic γ-counter (Wizard 3″). At 90 min after tracer injection, blood, brain, heart, plasma, muscle, lungs, thymus, bone, spleen, nose, kidney, liver, and intestines were collected after sacrifice and dissection. Radioactivity within tissue samples was counted using the automatic γ-counter, and counts per minute for each tissue sample were normalized to the total injected dose of radioactivity and weight of tissue to give the radioactivity uptake as percentage injected dose (%ID)/g. Before sacrifice, urine was collected and analyzed by radio–thin-layer chromatography using the same quality control method as during synthesis to determine stability (27).
Data Analysis
Uptake was compared between U87 (CXCR4 low-expressing) and U87.CXCR4 (human CXCR4 high-expressing) tumors via an unpaired 2-tailed t test. Tumor and liver uptake was compared in blocked and nonblocked animals via an unpaired 1-tailed t test. P values of less than 0.05 were considered to be statistically significant.
RESULTS
Synthesis and Characterization
The nonradioactive analog CuCB-bicyclam was formed by reacting the previously described CB-bicyclam with a stoichiometric amount of copper(II) acetate to form the mono-copper derivative, after which analytically pure samples were obtained using size-exclusion chromatography (Supplemental Fig. 2) (39). The in vitro binding affinity of the mono-copper complex to both human and murine CXCR4 was determined via a CXCL12-stimulated calcium flux signaling assay, and the half-maximal inhibitory concentration was determined to be 8 nM (human CXCR4), whereas the free chelator precursor CB-bicyclam displayed no measurable activity in this assay (>5 μM) (Supplemental Fig. 4). Similar results were observed for murine CXCR4, with a half-maximal inhibitory concentration of 2 nM for CuCB-bicyclam in a murine CXCL12 binding assay (Supplemental Fig. 5).
Radiochemistry, Purification, and Stability
64Cu-CuCB-bicyclam was radiochemically synthesized by heating CB-bicyclam with preformed 64Cu-Cu(OAc)2 and monitoring via radio–thin-layer chromatography (Supplemental Fig. 3). On full incorporation (40–50 min), the crude complex was purified of excess CB-bicyclam using semipreparative high-performance liquid chromatography to give decay-corrected isolated radiochemical yields of 69% ± 5% with a total synthesis time of 120 min and a specific activity of 7.15 GBq/μmol. The LogP value was −2.38 ± 0.18. 64Cu-CuAMD3100 was synthesized and analyzed following previously published procedures (28). In vitro acid stability was measured to indicate comparative stability between tracers by incubation in 6 M HClO4 for 3 h at 37°C. Radio–thin-layer chromatography methods identical to those used to analyze the radiosynthesis process were applied, and stability values of 92% ± 3% and 9% ± 5% were determined for 64Cu-CuCB-bicyclam and 64Cu-CuAMD3100, respectively.
Receptor Expression and 64Cu-CuCB-Bicyclam Radioligand Binding In Vitro
CXCR4 surface expression levels were determined by fluorescence-activated cell sorting for U87 and U87.CXCR4 cells (ca. 4,000 vs. 148,000 receptors per cell, respectively) (Fig. 1). Radioligand binding experiments showed ratios similar to the expression level for binding of the tracer (0.66% ± 0.56% and 20.22% ± 5.78% applied dose for U87 and U87.CXCR4 cells, respectively) (Fig. 1). Preincubation of U87.CXCR4 cells with Cu2CB-bicyclam (half-maximal inhibitory concentration, 3 nM) decreased binding of 64Cu-CuCB-bicyclam by 89%, consistent with high specific uptake.
PET/CT Studies on Tumor-Bearing Mice
Dynamic PET/CT studies performed on U87 and U87.CXCR4 xenograft–bearing mice demonstrated significantly higher uptake in U87.CXCR4 tumors than in U87 tumors after administration of 64Cu-CuCB-bicyclam (9.6 ± 0.7 MBq), with an SUVmax of 7.36% ± 1.77% versus 0.80% ± 0.14% at 80–90 min after injection (Fig. 2). Tumor-to-muscle ratios at 90 min after injection were 23.6 ± 2.7 for U87.CXCR4 tumors and 3.0 ± 0.5 for U87 tumors. Uptake was also seen in the kidneys and liver. Injection of a 5 mg/kg dose of Cu2CB-bicyclam 60 min before scanning reduced U87.CXCR4 tumor and liver uptake by more than 90%. Supplemental Figures 6 and 7 present all scan data and further analysis.
In vivo PET/CT evaluation of 64Cu-CuCB-bicyclam. (A) Fused PET/CT maximum-intensity projections at 70–90 min after injection from CD1 mice bearing U87-CXCR4 (n = 3), U87 (n = 3), and U87.CXCR4 block (n = 2). Animals were injected with 9.6 ± 0.7 MBq of 64Cu-CuCB-bicyclam. A 5 mg/kg blocking dose of Cu2CB-bicyclam was given 1 h before 64Cu-CuCB-bicyclam injection where indicated. Arrows denote position of tumor and liver. (B) Dynamic tumor time–activity curve for showing uptake of 64Cu-CuCB-bicyclam in tumor (top) and liver (bottom) during 90 min. Data represent mean uptake of 2–3 animals ± SEM. *U87.CXCR4 vs. U87, P < 0.01. **Block vs. naïve, P < 0.01. Supplemental Figures 6 and 7 show all images at 70–90 min and time–activity curves for bladder, kidney, and vena cava.
Ex Vivo Analysis of CXCR4 Expression
CXCR4 expression in U87.CXCR4 and U87 tumors and murine liver was confirmed ex vivo by a combination of Western blot, immunohistochemistry, and quantitative polymerase chain reaction (Supplemental Fig. 8; Supplemental Table 1).
Biodistribution and Tracer Stability Studies in Immunocompetent Mice
Biodistribution studies were performed using CD1-IGS mice at 90 min to determine the effect of a functional immune system on tracer distribution (Fig. 3). Low uptake (<2 %ID/g) was noted in all organs except kidney (8.11 %ID/g) and organs known to express CXCR4 (spleen [4.71 %ID/g], lung [2.89 %ID/g], and liver [13.77 %ID/g]). Urine was analyzed from a separate cohort of animals injected with 64Cu-CuCB-bicyclam and 64Cu-CuAMD3100 before sacrifice. 64Cu-CuCB-bicyclam was more than 90% intact, whereas less than 10% of 64Cu-CuAMD3100 remained intact in the urine.
Assessment of in vivo uptake of 64Cu-CuCB-bicyclam after blocking with range of CXCR4 antagonists. Biodistribution of selected tissues from CD1-IGS mice given 5 mg/kg dose of Cu2CB-bicyclam, CuCB-bicyclam, AMD3100, or AMD3465 60 min (top) or 12 h (bottom) before 64Cu-CuCB-bicyclam injection. Tissue was harvested 90 min after tracer administration (n = 4). *Block vs. naïve, P < 0.05.
Using murine liver CXCR4-dependent uptake of 64Cu-CuCB-bicyclam as a readout, the in vivo affinities or residence times of Cu2CB-bicyclam, CuCB-bicyclam, AMD3100, and AMD3465 were compared. Doses were administered 60 min before tracer administration, and animals were sacrificed at several time points after injection to provide tissue for biodistribution. Consistent with PET imaging in CD1-Nude mice, predosing with the Cu2CB-bicyclam significantly reduced tracer uptake at 60 min in liver (3.40 vs. 13.77 %ID/g, P < 0.05), spleen (0.48 vs. 4.71 %ID/g, P < 0.01), and lungs (1.06 vs. 2.89 %ID/g, P < 0.05) but not 12 h; similar results were seen using CuCB-bicyclam. No significant decrease in liver uptake was seen when animals were preinjected with AMD3100 or AMD3465 at any time point.
DISCUSSION
This study reports a high-affinity CXCR4-specific PET imaging agent (64Cu-CuCB-bicyclam) that binds selectively and with high stability to both the human and the murine variants of the CXCR4 receptor. Although 64Cu-CuCB-bicyclam has potential as a clinical diagnostic PET imaging agent for oncology, it also represents a useful tool for the interrogation of CXCR4 expression levels over time in genetically engineered or immunocompetent preclinical models, useful in the investigation of CXCR4-targeting synergy with immunomodulatory therapies (49).
Unlike AMD3100 and AMD3465, the labeling precursor CB-bicyclam has CXCR4 affinity of more than 5 μM because alkylation of the secondary amines disrupts the hydrogen-bonding potential. The mono-copper species CuCB-bicyclam has a higher affinity to the CXCR4 receptor than either AMD3100 or CuAMD3100 but lower than our previously described bis-copper species (Cu2CB-bicyclam). However, the affinity is likely to be sufficient for imaging studies, and 64Cu-CuCB-bicyclam has the advantage of a precursor with more than 100-fold lower receptor affinity. 64Cu radiolabeling was performed using the minimum concentration required for full radiometal incorporation. We chose to purify 64Cu-CuCB-bicyclam from unreacted CB-bicyclam using semipreparative high-performance liquid chromatography for these experiments.
Radioligand binding experiments using U87.CXCR4 and U87 cells showed specific binding, with binding correlating well with quantified surface expression levels of CXCR4. In addition, in vitro binding could be blocked using the higher-affinity complex Cu2CB-bicyclam. In vivo PET/CT imaging using U87.CXCR4 and U87 tumor–bearing mice mirrored the in vitro binding results, with selective high uptake in U87.CXCR4 tumors at 90 min after injection. In vivo specificity was confirmed using a blocking dose of the higher-affinity Cu2CB-bicyclam, which reduced U87.CXCR4 tumor uptake by more than 90%.
As seen in previous studies using 64Cu-labeled macrocycles for CXCR4 imaging, moderate liver uptake was seen on PET imaging using 64Cu-CuCB-bicyclam. This uptake was reduced after blocking with Cu2CB-bicyclam (5 mg/kg) to an extent comparable to that of U87.CXCR4 tumors, suggesting specific binding to CXCR4. This finding is comparable to previous studies in which very high doses of AMD3100 (50 mg/kg) were used for blocking (33). Murine liver expression of CXCR4 was confirmed after ex vivo disaggregation of tumor or liver at the protein level via fluorescence-activated cell sorting and immunohistochemistry. Further experiments demonstrated that more than 90% of urine radioactivity was intact 64Cu-CuCB-bicyclam, whereas most activity from 64Cu-CuAMD3100–injected animals was in the form of free 64Cu ions. Acid stability assays (6 M HClO4) showed similar profiles for stability. These data are consistent with in vitro studies, demonstrating that cyclam copper(II) complexes that do not possess either structural reinforcement or coordinating arms have rapid binding kinetics but low stability in competition experiments (31,50,51). These data confirm not only the insufficient stability of any nonbridged or nonfunctionalized cyclam-based copper(II) complex for in vivo PET imaging applications but also the higher stability of cross-bridged cyclam structures.
Further biodistribution studies were performed using immunocompetent mice to study the effect of blocking with various CXCR4 antagonists on the uptake of 64Cu-CuCB-bicyclam. As expected, liver uptake of the tracer could be blocked by administration of either the higher-affinity Cu2CB-bicyclam or CuCB-bicyclam at 60 min before tracer administration. However, tracer uptake could not be blocked by the lower-affinity CXCR4 antagonists AMD3100 and AMD3465 (Fig. 3; Supplemental Table 2). This observation could be explained by shorter (i.e., <60 min) receptor residence times for the unbridged cyclam compounds or displacement of AMD3100 or AMD3465 from CXCR4 by the higher-affinity 64Cu-CuCB-bicyclam compound (30). Although the development of therapeutic agents targeting CXCR4 was not the focus of this study, the high stability, high affinity, and appropriate biodistribution of 64Cu-CuCB-bicyclam create interest in exploring the identical radioactive and nonradioactive analogs as an imaging–chemotherapeutic pair. The 67Cu derivative is of interest for investigating a radionuclide theranostic approach (52), assuming that the CXCR4-related liver uptake is lower in humans than in mice.
CONCLUSION
In this study we have developed a 64Cu-labeled configurationally restricted azamacrocyclic CXCR4 antagonist with high affinity and stability and with favorable biodistribution characteristics. To our knowledge, 64Cu-CuCB-bicyclam is the first CXCR4-targeted PET imaging agent that has suitable characteristics for routine imaging studies and binds to both the human and the murine CXCR4 receptor variants. This agent allows wider preclinical assessment of the role of CXCR4 in syngeneic tumor models and immunocompetent models and will be useful for translation to other animal models of disease that require imaging of the endogenous CXCR4 receptor.
DISCLOSURE
Funding was provided by the Daisy Appeal Charity (grant DAhul0211 and BPB fellowship) and by Yorkshire Cancer Research (HEND376). The University of Hull provided studentship funding to Rhiannon Lee and Cecilia Miranda. No other potential conflict of interest relevant to this article was reported.
KEY POINTS
QUESTION: The goal of this study was to develop a configurationally restricted azamacrocyclic compound as a high-affinity and high-stability 64Cu radiotracer for imaging CXCR4 chemokine receptor expression in mice and humans.
PERTINENT FINDINGS: The radiotracer 64Cu-CuCB-bicyclam binds with high affinity to both murine and human CXCR4. The compound retains the 64Cu radiolabel and is excreted intact, with blocking studies in vivo showing significantly reduced uptake in liver, in spleen, and in human CXCR4–expressing tumor.
IMPLICATIONS FOR PATIENT CARE: 64Cu-CuCB-bicyclam is an appropriate tracer for studies of murine CXCR4 expression in mouse models and can be translated for use in humans.
Acknowledgments
We gratefully acknowledge Dr. Assem Allam and his family for the generous donation to help found the PET Research Centre at the University of Hull. Mass spectrometry data were acquired at the EPSRC U.K. National Mass Spectrometry Facility at Swansea University.
Footnotes
Published online Jun. 14, 2019.
- © 2020 by the Society of Nuclear Medicine and Molecular Imaging.
REFERENCES
- Received for publication September 20, 2018.
- Accepted for publication June 3, 2019.
Jump to section
Related Articles
Cited By...
- No citing articles found.