2858
`
`Chem. Rev. 2010, 110, 2858–2902
`
`Coordinating Radiometals of Copper, Gallium, Indium, Yttrium, and Zirconium
`for PET and SPECT Imaging of Disease
`
`Thaddeus J. Wadas,*,† Edward H. Wong,‡,§ Gary R. Weisman,‡,§ and Carolyn J. Anderson*,†
`
`Mallinckrodt Institute of Radiology, Washington University School of Medicine, 510 S. Kingshighway Blvd., Campus Box 8225 St. Louis,
`Missouri 63110, and Department of Chemistry, University of New Hampshire, Durham, New Hampshire 03824-3598
`
`Received September 29, 2009
`
`Imaging Hypoxia and Perfusion
`4.4.
`5. Conclusion
`6. Glossary
`7. Acknowledgments
`8. References
`
`2892
`2893
`2893
`2893
`2893
`
`2858
`2860
`
`Contents
`1.
`Introduction
`2. The Coordination Chemistry of Cu, Ga, Y, In, and
`Zr
`2860
`2.1. General Considerations
`2862
`2.2. Aqueous Copper Coordination Chemistry
`2863
`2.3. Copper(II) Complexes of Selected Chelators
`2863
`2.3.1. Acyclic Tetradentate Chelators
`2864
`2.3.2. Acyclic Hexadentate Chelators
`2865
`2.3.3. Macrocyclic Chelators
`2869
`2.4. Aqueous Gallium(III) Coordination Chemistry
`2869
`2.4.1. Tetradentate Ligands
`2870
`2.4.2. Hexadentate Ligands
`2872
`2.5. Aqueous Indium(III) Coordination Chemistry
`2872
`2.5.1. Tetradentate Chelators
`2873
`2.5.2. Hexa- to Octadentate Chelators
`2875
`2.6. Aqueous Yttrium(III) Coordination Chemistry
`2.7. Aqueous Zirconium(IV) Coordination Chemistry 2877
`2.8. Summary
`2877
`3. Radioisotope Production
`2878
`3.1. Production of Copper Radiometals
`2878
`3.2. Production of Gallium Radiometals
`2879
`3.3. Production of Indium Radiometals
`2879
`3.4. Production of Yttrium Radiometals
`2880
`3.5. Production of Zirconium Radiometals
`2881
`4. Applications of Zr, Y, Ga, In, and Cu
`2881
`Radiopharmaceuticals
`4.1. Oncology
`4.1.1.
`Integrin Imaging
`4.1.2. Somatostatin Receptor Imaging
`4.1.3. HER-2/neu Receptor Imaging
`4.1.4. Gastrin Releasing Peptide Receptor
`Imaging
`4.1.5. Epidermal Growth Factor Receptor Imaging 2888
`4.1.6. Vascular Endothelial Growth Factor
`2888
`Receptor Imaging
`4.1.7. Melanoma Imaging
`4.2.
`Imaging Gene Expression
`4.3.
`Imaging Inflammation and Infection
`
`1. Introduction
`Molecular imaging is the visualization, characterization,
`and measurement of biological processes at the molecular
`and cellular levels in humans and other living systems.
`Molecular imaging agents are probes used to visualize,
`characterize, and measure biological processes in living
`systems. These two definitions were put forth by the Society
`of Nuclear Medicine (SNM) in 2007 as a way to capture
`the interdisciplinary nature of this relatively new field. The
`emergence of molecular imaging as a scientific discipline is
`a result of advances in chemistry, biology, physics, and
`engineering, and the application of imaging probes and
`technologies has reshaped the philosophy of drug discovery
`in the pharmaceutical sciences by providing more cost-
`effective ways to evaluate the efficacy of a drug candidate
`and allow pharmaceutical companies to reduce the time it
`takes to introduce new therapeutics to the marketplace.
`Finally, the impact of molecular imaging on clinical medicine
`has been extensive since it allows a physician to diagnose a
`patient’s illness, prescribe treatment, and monitor the efficacy
`of that treatment noninvasively.
`Single-photon emission computed tomography (SPECT)
`and positron emission tomography (PET) were the first
`molecular imaging modalities used clinically. SPECT re-
`quires the use of a contrast agent labeled with a γ-emitting
`radionuclide, which should have an ideal γ energy of
`100-250 keV. These γ rays are recorded by the detectors
`of a dedicated γ camera or SPECT instrument and after signal
`processing can be converted into an image identifying the
`localization of the radiotracer. PET requires the injected
`radiopharmaceutical to be labeled with a positron-emitting
`radionuclide. As the radionuclide decays, it ejects a positron
`from its nucleus, which travels a short distance before being
`annihilated with an electron to release two 511 keV γ rays
`180° apart that are detected by the PET scanner (Figure 1).
`After sufficient acquisition time, the data are reconstructed
`using computer-based algorithms to yield images of the
`* Corresponding authors: Carolyn J. Anderson, phone 314.362.8427, fax
`314.362.9940, e-mail andersoncj@wustl.edu; Thaddeus J. Wadas, phone
`radiotracer’s location within the organism. Compared with
`314.362.8441, fax 314.362.9940, e-mail wadast@wustl.edu.
`SPECT, PET has greater advantages with respect to sensitiv-
`† Washington University School of Medicine.
`‡ University of New Hampshire.
`ity and resolution and has been gaining in clinical popularity,
`§ Contact information: Edward H. Wong, phone 603-862-1788, fax 603-
`with the number of PET-based studies expected to reach 3.2
`862-4278, e-mail ehw@cisunix.unh.edu; Gary R. Weisman, phone 603-
`million by 2010.1 While SPECT and PET technologies have
`862-2304, fax 603-862-4278, e-mail gary.weisman@unh.edu.
`10.1021/cr900325h  2010 American Chemical Society
`Published on Web 04/23/2010
`
`2881
`2881
`2883
`2886
`2887
`
`2889
`2890
`2891
`
`Petitioner GE Healthcare – Ex. 1012, p. 2858
`
`

`

`Coordinating Radiometals for PET and SPECT Imaging
`
`Chemical Reviews, 2010, Vol. 110, No. 5 2859
`
`Thaddeus J. Wadas was born in Nanticoke, PA, in 1974. He received his
`B.S. degree in Biology from King’s College, Wilkes-Barre, PA, in 1996
`and, while working in the environmental science industry, completed his
`second major in Chemistry in 1998. After completing his second major,
`he pursued graduate studies at the University of Rochester, Rochester,
`NY, where he received his M.S. and Ph.D. degrees in Chemistry under
`the supervision of Richard Eisenberg. His Ph.D. work focused on the
`synthesis and characterization of luminescent Pt(II) acetylide complexes
`for photoinduced charge transfer and light-to-chemical energy conversion.
`On completion of his Ph.D., he moved to the Washington University School
`of Medicine in St. Louis, MO, to pursue postdoctoral studies with Carolyn
`Anderson and develop targeted radiopharmaceuticals for diagnostic
`imaging and radiotherapy. In 2005, he was the recipient of a National
`Institutes of Health National Research Service Award (NRSA) Fellowship
`to study bone metastasis imaging with copper-64-labeled peptides, and
`in 2009 he was promoted to the position of Instructor at the School of
`Medicine. His current
`research interests include the application of
`combinatorial display methods to radiopharmaceutical development and
`understanding the respective roles gadolinium-based contrast agents and
`renal insufficiency play in the development of nephrogenic systemic fibrosis.
`
`Edward H. Wong was born in Wuhan, China, in 1946 but was raised in
`Hong Kong where he discovered the joy of mixing chemicals at St. Louis
`High School. He then majored in chemistry at the University of California
`at Berkeley and obtained his Ph.D. doing boron hydride synthesis with
`William N. Lipscomb at Harvard University in 1974. After a postdoctoral
`stint with M. Frederick Hawthorne at the University of California, Los
`Angeles, he began his academic career at Fordham University in 1976
`before moving to the University of New Hampshire in 1978. He has
`performed research in main group boron and phosphorus chemistry as
`well as metal-phosphine coordination chemistry. In recent years, with
`his colleague Gary Weisman he has focused on the coordination chemistry
`of cross-bridged tetraamine macrocycles. Together they have also explored
`applications of these chelators in copper-based radiopharmaceuticals with
`Carolyn Anderson and her research group.
`
`been around for decades, their use remained limited because
`of the limited availability of relevant isotopes, which had to
`be produced in nuclear reactors or particle accelerators.
`However, the introduction of the small biomedical cyclotron,
`
`Gary R. Weisman was born in Mason, Ohio, in 1949, receiving his primary
`and secondary education in the public school system there. He was
`interested in chemistry from a young age, working with his cousin Thomas
`J. Richardson in their substantial home laboratories. He earned his B.S.
`in Chemistry with Distinction at the University of Kentucky in 1971, carrying
`out research with Robert D. Guthrie. At the University of Wisconsins
`Madison, he worked on conformational analysis of hydrazines and their
`radical cations under the mentorship of Stephen F. Nelsen, earning the
`Ph.D. in Organic Chemistry in 1976. After a postdoctoral stint with Donald
`J. Cram at the University of California, Los Angeles, in 1976-1977, he
`started his independent academic career at
`the University of New
`Hampshire, where he has enjoyed both teaching and research. He was
`Gloria G. and Robert E. Lyle Professor at UNH from 2005 to 2009 and
`has been the recipient of Outstanding Teaching awards in 1995 and 2009.
`He has been a visiting professor at the University of Wisconsin (1986),
`University of Bristol (1987, 1998), University of Melbourne (2005), and
`Australian National University (2005). He was a Wilsmore Fellow at the
`University of Melbourne in 2002. His research interests are in both physical
`organic and synthetic organic chemistry, with special emphasis on
`stereochemical aspects. His recent research has centered on the chemistry
`of polyaza molecules,
`ligand design and synthesis, and biomedical
`applications of coordination complexes. He has enjoyed a productive
`collaboration with Edward H. Wong for almost two decades and more
`recently with Carolyn J. Anderson.
`
`the self-contained radionuclide generator, and the dedicated
`small animal or clinical SPECT and PET scanners to
`hospitals and research facilities has increased the demand
`for SPECT and PET isotopes.
`Traditional PET isotopes such as 18F, 15O, 13N, and 11C
`have been developed for incorporation into small molecules,
`but due to their often lengthy radiosyntheses, short half-lives,
`and rapid clearance, only early time points were available
`for imaging, leaving the investigation of biological processes,
`which occur over the duration of hours or days, difficult to
`explore. With the continuing development of biological
`targeting agents such as proteins, peptides, antibodies and
`nanoparticles, which demonstrate a range of biological half-
`lives, a need arose to produce new radionuclides with half-
`lives complementary to their biological properties. As a
`result, the production and radiochemistry of radiometals such
`as Zr, Y, In, Ga, and Cu have been investigated as
`radionuclide labels for biomolecules since they have the
`potential to combine their favorable decay characteristics with
`the biological characteristics of the targeting molecule to
`become a useful radiopharmaceutical (Tables 1 and 2).2
`The number of papers published describing the production
`or use of these radiometals continues to expand rapidly, and
`in recognition of this fact, the authors have attempted to
`present a comprehensive review of this literature as it relates
`to the production, ligand development, and radiopharma-
`ceutical applications of radiometals (excluding 99mTc) since
`1999. While numerous reviews have appeared describing
`
`Petitioner GE Healthcare – Ex. 1012, p. 2859
`
`

`

`2860 Chemical Reviews, 2010, Vol. 110, No. 5
`
`Wadas et al.
`
`survey spanning 50 years of scientific discovery. To ac-
`complish this goal, this review has been organized into three
`sections: the first section discusses the coordination chemistry
`of the metal ions Zr, Y, In, Ga, and Cu and their chelators
`in the context of radiopharmaceutical development;
`the
`second section describes the methods used to produce Zr,
`Y, In, Ga, and Cu radioisotopes; and the final section
`describes the application of these radiometals in diagnostic
`imaging and radiotherapy.
`
`2. The Coordination Chemistry of Cu, Ga, Y, In,
`and Zr
`
`2.1. General Considerations
`The development of metal-based radiopharmaceuticals
`represents a dynamic and rapidly growing research area that
`requires an intimate knowledge of metal coordination
`chemistry and ligand design. This section of the review
`covers general considerations regarding the parameters that
`are important in developing stable, kinetically inert radio-
`metal complexes that can be incorporated into radiophar-
`maceuticals. Additionally, the aqueous coordination chem-
`istry of these metals and their coordination complexes that
`are most relevant to radiopharmaceutical development are
`discussed below.
`Relevant properties in aqueous solution of the five metal
`cations covered in this review are presented in Table 3. The
`acidic cations Ga(III), In(III), and especially Zr(IV) present
`precipitation problems at neutral pH in the absence of suitable
`complex formation. In terms of plausible aqueous redox
`processes relevant to radiopharmaceutical applications, only
`Cu(II) and its complexes are susceptible to reduction
`chemistry, although the possibility of an ascorbic acid
`reduction of a 89Zr(IV) complex has been postulated.19 Based
`on Pearson’s hard-soft acid-base theory, the tetravalent
`Zr(IV) is an extremely hard acidic cation, followed by Y(III),
`Ga(III), and In(III). The Cu(II) cation is considered a
`borderline acid.
`
`Carolyn Anderson was born in Superior, WI, in 1962, and remained there
`throughout her school years. In 1985, she graduated Summa Cum Laude
`with a B.S. in Chemistry from the University of WisconsinsSuperior. In
`1984, she received a fellowship to attend the Summer School in Nuclear
`Chemistry at San Jose State University, and this is where her interest in
`nuclear and radiochemistry began. She pursued her Ph.D. in Inorganic
`Chemistry with Prof. Gregory R. Choppin at Florida State University,
`studying the electrochemistry and spectroscopy of uranium complexes in
`room-temperature molten salts. On completion of her Ph.D. in 1990, she
`moved to Washington University School of Medicine (WUSM) in St. Louis,
`MO, to carry out postdoctoral research with Prof. Michael J. Welch in the
`development of radiopharmaceuticals for PET imaging. In 1993, she was
`promoted to Assistant Professor of Radiology, and she is currently
`Professor in the departments of Radiology, Biochemistry & Molecular
`Biophysics, and Chemistry. Her research interests include the development
`of radiometal-labeled tumor receptor-based radiopharmaceuticals for PET
`imaging and targeted radiotherapy of cancer and cancer metastasis. She
`greatly enjoys the productive collaboration with Edward Wong and Gary
`Weisman on the development of novel chelation systems for attaching metal
`radionuclides to biomolecules for nuclear medicine imaging applications.
`
`certain aspects of the production, coordination chemistry, or
`application of these radiometals,2-18 very few exhaustive
`reviews have been published.10,12 Additionally, this review
`has been written to be used as an individual resource or as
`a companion resource to the review written by Anderson
`and Welch in 1999.12 Together, they provide a literature
`
`Table 1. γ- and (cid:2)-Emitting Radiometals
`Isotope
`production methods
`t1/2 (h)
`67Cu
`accelerator 67Zn(n,p)
`62.01
`67Ga
`78.26
`cyclotron
`90Y
`90Sr/90Y generator
`64.06
`111In
`cyclotron, 111Cd(p,n)111n
`67.9
`
`Table 2. Positron-Emitting Radiometals
`isotope
`methods of production
`t1/2 (h)
`60Cu
`cyclotron, 60Ni(p,n)60Cu
`0.4
`
`61Cu
`
`62Cu
`
`64Cu
`
`66Ga
`
`68Ga
`
`86Y
`
`89Zr
`
`3.3
`
`0.16
`
`12.7
`
`9.5
`
`1.1
`
`14.7
`
`78.5
`
`cyclotron, 61Ni(p,n)61Cu
`
`62Zn/62Cu generator
`
`cyclotron, 64Ni(p,n)64Cu
`
`cyclotron, 63Cu(R,nγ)66Ga
`
`68Ge/68Ga generator
`
`cyclotron, 86Sr(p,n)86Y
`
`89Y(p,n)89Zr
`
`decay mode
`(cid:2)- (100%)
`EC (100%)
`(cid:2)- (72%)
`EC (100%)
`
`Eγ (keV)
`91, 93, 185
`91, 93, 185, 296, 388
`
`245, 172
`
`E(cid:2) (keV)
`577, 484, 395
`
`2288
`
`ref
`578
`578
`578
`578
`
`decay mode
`(cid:2)+ (93%)
`EC (7%)
`(cid:2)+ (62%)
`EC (38%)
`(cid:2)+ (98%)
`EC (2%)
`(cid:2)+ 19(%)
`EC (41%)
`(cid:2)- (40%)
`(cid:2)+ (56%)
`EC (44%)
`(cid:2)+ (90%)
`EC (10%)
`(cid:2)+ (33%)
`EC (66%)
`
`(cid:2)+ (22.7%)
`EC (77%)
`
`E(cid:2)+ (keV)
`3920, 3000
`2000
`1220, 1150
`940, 560
`2910
`
`656
`
`4150, 935
`
`1880, 770
`2335, 2019
`1603, 1248
`1043
`897
`909, 1675, 1713, 1744
`
`ref
`578
`
`578
`
`578
`
`578
`
`578
`
`578
`
`578
`
`208, 578
`
`Petitioner GE Healthcare – Ex. 1012, p. 2860
`
`

`

`Coordinating Radiometals for PET and SPECT Imaging
`
`Chemical Reviews, 2010, Vol. 110, No. 5 2861
`
`Figure 1. Cartoon depicting the fundamental principle of positron emission tomography (PET). As the targeting group interacts with the
`cell surface receptor, the positron-emitting radiometal decays by ejecting (cid:2)+ particles from its nucleus. After traveling a short distance in
`the electron-rich tissue, the positron recombines with an electron in a process called annihilation. During annihilation, the mass of the
`positron and electron are converted into two high-energy photons (511 keV γ rays), which are released approximately 180° apart to ensure
`that energy and momentum are conserved. Although attenuation is possible, these two γ rays are usually energetic enough to escape the
`organism and be collected by the detectors of a PET scanner.
`
`Since the preponderance of radiometal complexes of note
`feature at least tetradentate ligands, we have restricted our
`
`cation/electron
`configuration
`Cu(II)/[Ar]3d9
`
`Ga(III)/[Ar]3d10
`
`In(III)/[Kr]4d10
`
`Y(III)/[Kr]
`
`Zr(IV)/[Kr]
`
`Table 3. Properties of Relevant Metal Cations
`ionic
`radiusa
`kexchange,c
`s-1
`(CN)
`pKa
`57 (4) 7.53 2 × 108
`65 (5)
`73 (6)
`47 (4) 2.6
`55 (5)
`62 (6)
`62 (4) 4.0
`80 (6)
`92 (8)
`90 (6) 7.7
`102 (8)
`108 (9)
`59 (4) 0.22
`72 (6)
`84 (8)
`89 (9)
`
`b
`
`Ered,d V,
`(acid)
`+0.34 (Cu0)
`+0.16 (CuI)
`7.6 × 102 -0.56 (Ga0)
`-0.65 (GaII)
`4.0 × 104 -0.34 (In0)
`-0.49 (InII)
`1.3 × 107 -2.37 (Y0)
`
`-1.54 (Zr0)
`
`hardness
`classification
`(IA)e
`borderline (2.68)
`
`hard (7.07)
`
`hard (6.30)
`
`hard (10.64)
`
`hard
`
`a Picometers.579 b As hydrated cation.580 c In H2O.581 d Versus NHE;
`ref 582, Table 6.2, p 267; ref 583, Appendix E. e IA ) EA/CA; refs 584
`and 585; ref 586, Table 2.3.
`
`discussion here to ligands with four or more donor sites
`coordinating the cation of interest. Rather than exhaustive
`coverage of all chelators of potential interest, we will discuss
`only selected representatives of the most-frequently reported
`ligands, especially those with more complete data of
`relevance. For the chosen representative chelators of each
`cation, we have listed available pertinent data on their
`denticity, coordination geometry, and thermodynamic stabil-
`ity. Where X-ray structural data are available, geometrical
`data on the coordination mode can provide useful insight
`into the “goodness of fit” for a specific cation-chelator
`pairing, the caveat being that actual solution structures or
`indeed number of species may be distinct from solid-state
`observations. For the four diamagnetic cations, solution NMR
`spectroscopic studies can be used to supplement X-ray data.
`Despite the difficulty of comparing stability constants of
`complex formation between ligands of different basicity and
`denticity, the listed log KML’s provide a convenient gauge
`of their relative affinities for a specific metal.
`For in ViVo applications, kinetic inertness of metal-chelator
`complexes or conjugates can be more relevant than thermo-
`dynamic stability.12,20,21 In general, acyclic chelator com-
`
`Petitioner GE Healthcare – Ex. 1012, p. 2861
`
`

`

`2862 Chemical Reviews, 2010, Vol. 110, No. 5
`
`Wadas et al.
`
`plexes are less kinetically inert than macrocyclic complexes
`of comparable stability.22-26 By the same token, acyclic
`chelators typically have faster metal-binding kinetics com-
`pared with their macrocyclic analogues, which can be a
`significant advantage for shorter-lived radiometals.27-30 There
`have been efforts to enhance the binding rate of macrocycles
`by incorporation of an acyclic polydentate pendant arm.31
`A variety of in Vitro assays of metal-chelator complex
`integrity can be found in the literature.32-35 A popular assay
`of aqueous kinetic inertness is acid decomplexation. This
`has some relevance in biological environments that are
`relatively acidic such as in hypoxic tissues and certain cell
`vesicles. However, the extremely high acidities, for example,
`1-5 M HCl, often required to decompose relatively inert
`complexes clearly have no parallel to any in ViVo conditions.
`Nor can such data be relied upon, without considerations of
`other factors, as the sole predictor of biological behavior.36
`Typically, the decomplexation of Cu(II) complexes is readily
`monitored through their electronic spectra. Demetalation of
`the diamagnetic Ga(III), In(III), Y(III), and Zr(IV) complexes
`can usually be followed by proton and 13C NMR spectros-
`copy in acidified D2O solutions. Where feasible, 71Ga, 115In,
`and 89Y NMR studies can also be undertaken.37-39 Although
`detailed mechanistic investigations are sometimes reported,
`more commonly only pseudo-first-order half-lives are re-
`ported, which should only be used to rank inertness
`qualitatively. Nonetheless, such data remain useful as a
`preliminary indicator of the in ViVo viability of specific metal-
`based radiopharmaceuticals.
`Competition or challenge assays of complexes of interest
`with excess biometals and biochelators are relevant since
`their typical concentrations are orders of magnitude higher
`than the radiolabeled complex’s, requiring high chelator
`selectivity for the radiometal. For example, copper homeo-
`stasis is tightly regulated in biology,40 and as a result, a
`variety of copper-binding biomolecules are present in ex-
`tracellular (serum albumin, ceruloplasmin, transcuprin, etc.)
`and intracellular (transporters, chaperones, metallothioneins,
`superoxide dismutase, cytochrome c oxidase, etc.) environ-
`ments.41-43 A viable Cu(II) chelator should therefore be both
`thermodynamically stable and kinetically inert to transche-
`lation challenges by these species. Highly charged cations
`like Y(III) and Zr(IV) may also have high affinity for bone
`tissues, while the avid Ga(III) binding of transferrin is
`well-established.44-46 Serum stability studies using radiometal-
`labeled chelator complexes or their bioconjugates are routinely
`used in inertness assays. These are readily monitored by radio-
`TLC, HPLC, and LC-MS techniques.47-49 In Vitro uptake
`studies using specific cell lines have also been carried out in
`many assays. While simulating extracellular environments to
`an extent, these studies cannot always accurately forecast in
`ViVo behavior. Ultimately, studies of animal biodistribution and
`bioclearance using radiometal-labeled complexes or bioconju-
`gates need to be carried out to obtain realistic data on their in
`ViVo performance.
`The following discussion of pertinent acyclic and macro-
`cyclic ligands and their specific metal coordination chemistry
`is organized according to their denticity. Most of these
`ligands have been designed to provide a minimum of four
`donor atoms, usually also incorporating anionic sites for
`charge balance (See Figures 2 and 3). While all are given
`numerical “L(number)” designations, many have been
`labeled additionally with their respective acronyms. Published
`X-ray crystal structures of Cu, In, Ga, Y, and Zr coordination
`
`complexes involving these ligands are also provided where
`appropriate. They were prepared from published CIF files
`using CrystalMaker 8.2 for Mac (CrystalMaker Software
`Ltd., Centre for Innovation & Enterprise, Oxford University
`Begbroke Science Park, Sandy Lane, Yarnton, Oxfordshire,
`OX5 1PF, UK; http://www.crystalmaker.com). Each atomic
`sphere is scaled to 0.4 times the covalent atomic radius, using
`the recently updated radii of Alvarez and co-workers.50 In
`addition to the labeled and uniquely colored metal atoms,
`common elements are color coded as follows: C ) gray, Cl
`) green, F ) light green, N ) blue, O ) red, P ) orange,
`and S ) yellow. Hydrogen atoms have been omitted from
`the structures for clarity.
`
`2.2. Aqueous Copper Coordination Chemistry
`While +1 and +3 oxidation states are both accessible for
`copper in the presence of suitable donors, 3d9 Cu(II) remains
`the predominant state for radiocopper chemistry in protic
`media. The aqueous cupric ion was long believed to have a
`tetragonally distorted hexa-aqua structure until a 2001 report
`suggested only five-coordination.51 Its water-exchange rate
`has been found to be very rapid compared with most common
`first-row transition metal cations and as a result it has
`relatively facile substitution chemistry despite having some
`crystal-field stabilization. This is usually ascribed to the
`Jahn-Teller distortion that elongates one or more of its
`coordinated ligands. Classified as a cation of borderline
`hardness, the high affinity of Cu(II) for borderline nitrogen
`donors is well-established. With a relatively small ionic
`radius of between 57 and 73 pm for coordination numbers
`4-6, it is particularly suitable for the formation of five-
`membered chelate rings; indeed the chelate effect is epito-
`mized in its ethylenediamine family of complexes.52 The
`popular use of polyazamacrocycles, especially cyclen and
`cyclam, for strong binding of Cu(II) is a consequence of the
`added advantage of the macrocyclic effect,53 as borne out
`by their extensive coordination literature.54-57
`The importance of in ViVo redox activation of metallodrugs
`incorporating Pt(IV), Ru(III), and Co(III) has received
`increasing attention.58-61 The role of bioreduction in copper
`radiopharmaceutical efficacy has been intensively studied in
`their thiosemicarbazone complexes, especially Cu-ATSM
`(L9).62-64 Convincing evidence for the formation and selec-
`tive retention/decomplexation of Cu(I)-intermediates from
`Cu(II) precursors in hypoxic tissues has been presented.65,66
`Whether Cu(II)/Cu(I) bioreduction is also a viable pathway
`for irreversible in ViVo radiocopper loss from other chelator
`complexes and their bioconjugates is an intriguing possibility.
`There is some compelling evidence for the deteriorated in
`ViVo performance of related Cu(II) complexes differing only
`in their reduction propensities. Specifically, the “long arm”
`dicarboxyethyl pendant-armed Cu(II) complex of cross-
`bridged cyclam has an Ered almost 400 mV higher (or more
`positive) than that its carboxymethyl-armed analogue, Cu-
`CB-TE2A (L57).67 The former has been found to exhibit
`significantly inferior bioclearance behavior despite very
`similar coordination geometry and acid-inertness. More
`structure-activity studies, including the consequence of
`protonation on reduction feasibility, are warranted. Most
`polyazamacrocyclic complexes of Cu(II), however, have
`rather negative reduction potentials that are well below the
`estimated -0.40 V (NHE) threshold for typical bioreductants.
`It should be further noted that an appropriate in ViVo donor
`able to alter the first or, perhaps even second coordination
`
`Petitioner GE Healthcare – Ex. 1012, p. 2862
`
`

`

`Coordinating Radiometals for PET and SPECT Imaging
`
`Chemical Reviews, 2010, Vol. 110, No. 5 2863
`
`Figure 2. Selected acyclic chelators.
`
`sphere around a metal cation can dramatically facilitate its
`redox processes. The relevance of this tuning of redox-active
`metal
`lability during biological
`iron transfer has been
`substantiated.68,69 Whether such ternary interactions can play
`a role in the reductive demetalation of thermodynamically
`stable Cu(II) complexes in ViVo has not been explored.
`
`2.3. Copper(II) Complexes of Selected Chelators
`The plasticity of the Cu(II) coordination geometry can be
`gleaned from a literature survey of 89 of its complexes with
`cyclen and cyclam derivatives.70 Coordination numbers (CN)
`ranging from 4 to 6 were found with geometries approximat-
`ing square planar, square pyramidal, trigonal bipyramidal,
`and octahedral. Tetradentate chelators are usually designed
`to cater to Cu(II)’s strong affinity for ligands favoring a
`square-planar geometry. Common donor sets include two
`amino or imino nitrogens combined with two charge-
`neutralizing anionic amido, oxo, or thiolato sites. These
`
`include numerous Schiff base or amino acid derived chela-
`tors. Full envelopment of Cu(II) in its maximum six-
`coordinate mode is much sought after. As a result, hexa-
`dentate chelators have become the most investigated in
`radiocopper chemistry. Popular scaffolds include triaza- or
`tetraazamacrocycles, especially TACN (L26), cyclen (L38),
`and cyclam (L48). Methodologies for selective attachment
`of appropriate pendant arms to their secondary amine
`nitrogen sites as well as to the carbon backbone have been
`developed.71-79 Resulting donor sets usually incorporate
`anionic carboxylate or thiolate sites to provide a medley of
`charge-neutralizing N3O3, N3S3, or N4O2 coordination spheres.
`Data for selected Cu(II)-chelator complexes are listed in
`Table 4.
`
`2.3.1. Acyclic Tetradentate Chelators
`A dimethyl ester of N,N′-ethylenediamine-di-L-cysteinato,
`EC (L5), was reacted with Cu(II), and the resulting complex
`
`Petitioner GE Healthcare – Ex. 1012, p. 2863
`
`

`

`2864 Chemical Reviews, 2010, Vol. 110, No. 5
`
`Wadas et al.
`
`Figure 3. Selected macrocyclic chelators.
`
`was structurally characterized and found to be substantially
`twisted (21°) from a square-planar geometry (Figure 4).80
`Bis(thiosemicarbazonato) complexes of cold and radio-
`Cu(II) have been intensely investigated for hypoxia imaging
`(Vide infra).62,63,81 A series of X-ray structures of these
`complexes have been determined, and near square-planar
`geometries were typically observed (e.g., Cu-GTS (Cu-L7)
`and Cu-ATSM (Cu-L9) in Figures 5 and 6). Alkylation at
`the backbone C atoms was found to increase the backbone
`C-C bond length and allow the metal to fit better into the
`ligand cavity with shorter Cu-S bonds.82 Their Cu(II)/Cu(I)
`reduction potentials have been shown to have significant
`bearing on their in ViVo biological behavior.62,63,66
`
`2.3.2. Acyclic Hexadentate Chelators
`The Cu(II)-EDTA (L10) structure has been reported to
`be a tetragonally distorted N2O4 octahedron along one
`O-Cu-O axis (Figure 7).83 A DTPA (L12) analogue also
`
`features a hexadentate chelator but with an N3O3 coordination
`environment (Figure 8).84
`Rigid bispidine (3,7-diazabicyclo[3.3.1]nonane) derivatives
`with two appended pyridyl functions have been shown to
`be tetradentate in five-coordinate Cu(II) complexes (Cu-L17,
`Figure 9).85 Variations with four pyridyl as well as two
`noncoordinating carboxylate groups for charge neutralization
`are hexadentate chelators, which were found to bind Cu(II)
`rapidly. An X-ray structure revealed a distorted octahedral
`N6 coordination mode (Cu-L18, Figure 10). This chelator
`was conjugated to bombesin, radiolabeled with 64Cu, and
`studied in rats.86
`Hexadentate ligands based on the 1,3,5-triaminocyclohex-
`ane backbone appended with three methylpyridines (TA-
`CHPYR, L19) have been investigated as radiocopper
`chelators.87,88 These form tetragonally distorted octahedral
`Cu(II) complexes (Figure 11).
`
`Petitioner GE Healthcare – Ex. 1012, p. 2864
`
`

`

`Coordinating Radiometals for PET and SPECT Imaging
`
`Chemical Reviews, 2010, Vol. 110, No. 5 2865
`
`chelator
`
`cation coordination geometry
`distorted square planar
`distorted square planar
`
`Table 4. Data on Selected Cu(II)-Chelator Complexes
`donor set
`(total CN)
`L6
`N2S2 (4)
`L9, ATSM
`N2S2 (4)
`L10, EDTA
`N2O4 (6)
`L12, DTPA
`N3O3 (6)
`L17
`N6 (6)
`L19, TACHPYR N6 (6)
`L29, NOTA
`N3O3 (6)
`L38, cyclen
`N4 (5)
`L34, L36
`N4 (5)
`L39, DOTA
`N4O2 (6)
`L46, DO2P
`N4O2 (6?)
`N4 (5-6)
`L48, cyclam
`
`distorted octahedron
`distorted octahedron
`distorted trigonal prism
`square pyramid
`distorted square pyramid
`distorted octahedron
`
`log KML
`
`a
`
`Ered or Ep
`
`b
`
`acid inertness, t1/2
`(conditions)
`
`ref
`
`-0.40 (q-rev)
`
`80
`81, 82
`587, 588
`589
`590
`88
`586, 591, 592
`112, 593
`99, 100
`145, 594, 595
`103
`593
`
`18.8, 19.2
`21.4
`16.3
`0.08 (q-rev)
`19.8, 21.6 ∼-0.70 (irrev)
`<3 min (5 M HCl, 30°)
`<3 min (5 M HCl, 30°)
`24.6
`∼-1.60b (irrev)
`8.3
`22.2, 22.7 ∼-0.74 (irrev)
`<3 min (5 M HCl, 90°)112
`28.7
`∼-0.48 (irrev)
`3.8 min (5 M HCl, 90°)112
`27.2
`square pyramid, tetragonally
`elongated octahedral
`21.1, 21.9 ∼-0.98 (irrev)
`<3 min (5 M HCl, 90°)112
`L49, TETA
`N4O2 (6)
`145, 594, 595
`distorted octahedron
`∼-0.45 (irrev)
`1.7 h (1 M HCl, 60°)
`L54, TE2P
`N4O2 (6)
`114
`26.5
`tetragonally distorted octahedron
`11.8 min (1 M HCl, 90°)
`-0.32 (q-rev)
`L56, CB-cyclam N4 (5)
`111, 112, 596
`27.1
`distorted square pyramid
`∼-0.72 (irrev)
`4.0 h (1 M HCl, 30°)
`L37, CB-DO2A N4O2
`21, 112
`distorted octahedron
`154 h (5 M HCl, 90°)
`-0.88 (q-rev)
`L57, CB-TE2A
`N4O2 (6)
`111, 112
`distorted octahedron
`∼-0.90 (irrev)13
`40 h (5 M HCl, 90°)13
`L62, DIAMSAR N6 (6)
`118
`distorted octahedron or trigonal prism
`a KML ) [ML]/[M][L]. b Ered ) reduction potential and Ep ) peak potential only, both vs NHE; (q-rev) ) quasi-reversible; (irrev) ) irreversible
`reduction.
`
`Figure 4. Cu-L6.
`
`Figure 5. Cu-GTS (L7).
`
`Figure 7. Cu-EDTA (L10).
`
`Figure 6. Cu-ATSM (L9).
`2.3.3. Macrocyclic Chelators
`The 14-membered N2S2 macrocycle L24 was found to
`form the most inert Cu(II) complex compared with other ring
`sizes.89 New N2S2 macrocycles with two appended car-
`boxymethyl arms (L25) have been synthesized and com-
`plexed with both C