Chem Soc Rev
`
`REVIEW ARTICLE
`
`Received 16th August 2013
`
`DOI: 10.1039/c3cs60304k
`
`www.rsc.org/csr
`
`1. Introduction
`
`Matching chelators to radiometals for
`radiopharmaceuticals
`
`Eric W. Price*ab and Chris Orvig*a
`
`Radiometals comprise many useful radioactive isotopes of various metallic elements. When properly
`harnessed, these have valuable emission properties that can be used for diagnostic imaging techniques,
`such as single photon emission computed tomography (SPECT, e.g. 67Ga, 99mTc, 111In, 177Lu) and positron
`emission tomography (PET, e.g. 68Ga, 64Cu, 44Sc, 86Y, 89Zr), as well as therapeutic applications (e.g. 47Sc,
`114mIn, 177Lu, 90Y, 212/213Bi, 212Pb, 225Ac, 186/188Re). A fundamental critical component of a radiometal-based
`radiopharmaceutical
`is the chelator, the ligand system that binds the radiometal
`ion in a tight stable
`coordination complex so that it can be properly directed to a desirable molecular target in vivo. This
`article is a guide for selecting the optimal match between chelator and radiometal for use in these
`systems. The article briefly introduces a selection of relevant and high impact radiometals, and their
`potential utility to the fields of radiochemistry, nuclear medicine, and molecular imaging. A description of
`radiometal-based radiopharmaceuticals is provided, and several key design considerations are discussed.
`The experimental methods by which chelators are assessed for their suitability with a variety of radiometal
`ions is explained, and a large selection of the most common and most promising chelators are evaluated
`and discussed for their potential use with a variety of radiometals. Comprehensive tables have been
`assembled to provide a convenient and accessible overview of the field of radiometal chelating agents.
`
`Radiometals are radioactive isotopes that can be harnessed for
`applications in medical diagnosis, as well as for cancer therapy.
`In order to apply these isotopes to specific biological applications,
`the ‘‘free’’ radiometal ions must be sequestered from aqueous
`solution using chelators (ligands) to obviate transchelation and
`hydrolysis. Chelators used for this application are typically
`covalently linked to a biologically active targeting molecule,
`making an active radiopharmaceutical agent. The chelator is
`used to tightly bind a radiometal ion so that when injected into a
`patient, the targeting molecule can deliver the isotope without
`any radiometal loss from the radiopharmaceutical, effectively
`supplying a site-specific radioactive source in vivo for imaging or
`therapy. A rapidly expanding number of radiometals are routi-
`nely produced, with a broad variety of half-lives, emission types,
`energies, and branching ratios (Table 1).1–4 The availability of a
`wide range of radiometal ions makes it possible to carefully pick
`the specific nuclear properties that are needed for a vast number
`of different applications. Some examples of radiometals that can
`be used for positron emission tomography (PET) imaging are
`
`a Medicinal Inorganic Chemistry Group, Department of Chemistry,
`University of British Columbia, 2036 Main Mall, Vancouver, British Columbia,
`Canada V6T 1Z1. E-mail: ericwprice@gmail.com, orvig@chem.ubc.ca
`b TRIUMF, 4004 Wesbrook Mall, Vancouver, British Columbia, Canada V6T 2A3
`
`68Ga, 64Cu, 86Y, 89Zr, and 44Sc, with PET imaging providing
`sensitive, quantitative, and non-invasive images of a variety of
`molecular processes and targets. Single photon emission computed
`tomography (SPECT) is an older and more ubiquitous imaging
`modality than PET, and, since its inception in the 1960s, 99mTc
`has been the workhorse isotope of SPECT. More recently, the
`radiometals 67Ga, 111In, and 177Lu have been increasingly used
`for SPECT imaging in chelator-based radiopharmaceuticals. For
`therapy applications, particle emitters such as 111In (Auger
`electron emitter), 90Y and 177Lu (b), and 225Ac, 212Pb, and
`213Bi (a), are being heavily investigated, typically in conjunction
`with antibody vectors (immunoconjugates) or peptides. Each
`radiometal ion has unique aqueous coordination chemistry
`properties; these must be properly attended to if these isotopes
`are to be safely harnessed for medical applications and use
`in vivo.
`The major difference between radioactive (‘‘hot’’) and non-
`radioactive (‘‘cold’’) metal ion chemistry is that radiochemistry
`is typically performed under extremely dilute conditions, with
`radiometal ions typically being utilized at nM to pM concentra-
`tions.5 It is also important to note that several of the elements
`being discussed have multiple radioactive isotopes that are
`for diagnostic or therapeutic purposes (e.g. 86/90Y,
`useful
`67/68Ga, 44/47Sc, 60/61/62/64Cu), and all isotopes of a given element
`have identical chemistry.6–11 This means that a single radio-
`pharmaceutical agent can be radiolabeled with different isotopes
`
`260 | Chem. Soc. Rev., 2014, 43, 260--290
`
`This journal is © The Royal Society of Chemistry 2014
`
`Cite this: Chem. Soc. Rev., 2014,
`43, 260
`
`Published on 30 October 2013. Downloaded on 1/27/2025 6:39:21 PM.
`
`View Article Online
`
`View Journal
`
` | View Issue
`
`Petitioner GE Healthcare – Ex. 1013, p. 260
`
`

`

`Review Article
`
`Chem Soc Rev
`
`of the same element (e.g. 86/90Y), and provide the same charge
`and physical properties, and therefore the same biological
`behavior and distribution in vivo.6–11 This class of radiophar-
`maceutical that utilizes two isotopes of the same element, such
`as 86Y for PET imaging and 90Y for therapy, has been referred to
`as a theranostic agent.12,13 An interesting note about 90Y is that
`as a b emitter it is possible to perform biodistribution,
`imaging, and dosimetry studies with its bremsstrahlung
`X-rays, but because the spatial resolution and image quality
`obtained is very poor an imaging surrogate is typically used (e.g.
`111In or 86Y).14 Alternatively, 90Y is unique and in some circum-
`stances can be used directly for PET imaging because it emits a
`very low abundance of positrons (0.003%), which can be used to
`collect imaging data superior to that obtained by 90Y brems-
`strahlung imaging.14
`
`2. Radiometal-based
`radiopharmaceutical design
`
`Ligands that are typically used to construct radiometal-based
`radiopharmaceuticals (not always with 99mTc) are bifunctional
`chelators (BFCs), which are simply chelators with reactive functional
`groups that can be covalently coupled (conjugated) to targeting
`vectors (e.g. peptides, nucleotides, antibodies, nanoparticles).
`Common bioconjugation techniques utilize functional groups
`such as carboxylic acids or activated esters (e.g. N-hydroxy-
`succinimide NHS-ester, tetrafluorophenyl TFP-ester) for amide
`couplings, isothiocyanates for thiourea couplings, and maleimides
`for thiol couplings (Fig. 1).17,18 Click chemistry is gaining popularity
`in bioconjugate chemistry, with both the traditional copper(I)
`catalyzed azide–alkyne Huisgen 1,3-dipolar cycloaddition ‘‘click’’
`
`reaction (forming a 1,2,3-triazole-ring linkage), or newer copper-
`free reactions like strain-promoted azide–alkyne cycloadditions
`(e.g. dibenzocyclooctyne/azide reaction) and Diels–Alder click
`reactions (e.g. transcyclooctene/1,2,4,5-tetrazine) (Fig. 1).19 It is
`interesting to note that the transcyclooctene/1,2,4,5-tetrazine
`copper-free click coupling displays remarkably fast reaction
`kinetics, allowing for novel applications like in vivo pre-
`targeting, where the click reaction can occur in vivo at very dilute
`concentrations.20–24 The modular design of BFC systems allows
`for a theoretically limitless number of different vectors to be
`conjugated, providing molecular targeting to a constantly
`increasing number of biological targets.
`The structure and physical properties of the radiometal–
`chelate complex have a large impact on the overall pharmaco-
`kinetic properties of a radiopharmaceutical, with many radio-
`metal complexes being very hydrophilic and subsequently
`leading to rapid renal excretion when attached to small vectors
`like peptides and nucleotides (less prominent with large
`B150 kDa antibodies).6–11,179 It has been observed in peptide-
`conjugates that keeping the radiometal ion and peptide con-
`stant, and changing only the chelator can have drastic effects on
`biodistributions.25 Radiometal-based radiopharmaceuticals con-
`tain many synthetically exchangeable components, which can be
`separated into different modules: the radiometal, which changes
`the radioactive emission properties and half-life (g for SPECT, b+
`for PET, and b or a particles, or Auger electrons, for therapy);
`the chelator, which must be carefully matched with the radio-
`metal for optimal stability; the BFC–vector conjugation method,
`for different types of bioconjugation reactions and linkages; and
`the vector/targeting moiety, which allows for the selection of any
`known molecular target for site-specific delivery of the radio-
`active ‘‘payload’’ (Fig. 2).
`
`Eric Price received his undergraduate education at the University of
`Victoria in Victoria, British Columbia, Canada, where he completed
`his honours BSc (2009) in Chemistry as a part of the CO-OP
`program, performing research both in industry and academia,
`including work at TRIUMF and an NSERC USRA research term
`with Dr Matthew Moffitt. In 2009 he began his PhD studies as an
`NSERC scholar in a joint project between Dr Chris Orvig at the
`University of British Columbia and Dr Michael J. Adam at TRIUMF,
`partially
`in collaboration with Nordion,
`researching new
`radiometal-based radiopharmaceuticals of 67/68Ga, 64Cu, 111In,
`177Lu, 89Zr, and 225Ac. He initiated, and has been pivotal in, an
`international research collaboration with Dr Jason S. Lewis and Dr
`Brian M. Zeglis, traveling to Memorial Sloan-Kettering Cancer
`Center in New York, USA, and has also collaborated with the BC
`Cancer Agency in Vancouver, as part of his PhD work. In 2013 he won the Berson–Yalow award from the Society of Nuclear Medicine and
`Molecular Imaging.
`Chris Orvig earned his Hons. BSc from McGill and his PhD as an NSERC of Canada scholar at MIT with Alan Davison, FRS. After
`postdoctoral fellowships with Kenneth N. Raymond at the University of California, Berkeley, and the late Colin J. L. Lock at McMaster
`University, he joined the University of British Columbia in 1984, where he is now Professor of Chemistry and Pharmaceutical Sciences.
`Orvig, a Fellow of the Royal Society of Canada, has received various teaching, research and service awards, and published more than 200
`research papers. He is a co-inventor on many issued patents, and a certified ski instructor.
`
`Eric W. Price and Chris Orvig
`
`This journal is © The Royal Society of Chemistry 2014
`
`Chem. Soc. Rev., 2014, 43, 260--290 | 261
`
`Published on 30 October 2013. Downloaded on 1/27/2025 6:39:21 PM.
`
`View Article Online
`
`Petitioner GE Healthcare – Ex. 1013, p. 261
`
`

`

`Chem Soc Rev
`
`Review Article
`
`Table 1 Properties of some popular radiometal isotopes, EC = electron capture; some low abundance emissions have been omitted for brevity1–4,15,16
`
`Isotope
`60Cu
`
`61Cu
`
`62Cu
`
`64Cu
`
`66Ga
`
`67Ga
`
`68Ga
`
`44Sc
`
`47Sc
`
`111In
`
`114mIn
`
`114In (daughter)
`
`177Lu
`
`86Y
`
`90Y
`
`89Zr
`
`212Bi
`
`213Bi
`
`212Pb (daughter is 212Bi)
`
`225Ac
`
`t1/2 (h)
`
`0.4
`
`3.3
`
`0.16
`
`12.7
`
`9.5
`
`78.2
`
`1.1
`
`3.9
`
`80.2
`
`67.2
`
`49.5 d
`
`73 s
`
`159.4
`
`14.7
`
`64.1
`
`78.5
`
`1.1
`
`0.76
`
`10.6
`
`240
`
`Decay mode
`b+ (93%)
`EC (7%)
`
`b+ (62%)
`EC (38%)
`
`b+ (98%)
`EC (2%)
`
`b+ (19%)
`EC (41%)
`b (40%)
`
`b+ (56%)
`EC (44%)
`
`E (keV)
`b+, 3920, 3000, 2000
`
`Production method
`Cyclotron, 60Ni(p,n)60Cu
`
`b+, 1220, 1150, 940, 560
`
`Cyclotron, 61Ni(p,n)61Cu
`
`b+, 2910
`
`b+, 656
`
`62Zn/62Cu generator
`
`Cyclotron, 64Ni(p,n)64Cu
`
`b+, 4150, 935
`
`Cyclotron, 63Cu(a,ng)66Ga
`
`EC (100%)
`
`g, 93, 184, 300
`
`Cyclotron, 68Zn(p,2n)67Ga
`
`b+ (90%)
`EC (10%)
`
`b+ (94%)
`EC (6%)
`b (100%)
`
`b+, 1880
`
`g, 1157
`b+, 1474
`
`g, 159
`b, 441, 600
`
`68Ge/68Ga generator
`
`44Ti/44Sc generator
`
`47Ti(n,p)47Sc
`
`EC (100%)
`
`g, 245, 172
`
`Cyclotron, 111Cd(p,n)111m,gIn
`
`EC (100%)
`b (100%)
`b (100%)
`
`b+ (33%)
`EC (66%)
`b (100%)
`
`b+ (23%)
`EC (77%)
`
`a (36%)
`b (64%)
`
`a (2.2%)
`b (97.8%)
`b (100%)
`
`g, 190
`b, 1989
`
`g, 112, 208
`b, 177, 385, 498
`
`b+, 1221
`
`b, 2280
`
`b+, 897
`
`a, 6050
`b, 6089
`
`a, 5549
`b, 5869
`
`a, 570
`
`Cyclotron, 114Cd(p,n)114mIn or 116Cd(p,3n)114mIn
`
`176Lu(n,g)177Lu
`
`Cyclotron, 86Sr(p,n)86Y
`
`90Zr(n,p)90Y
`
`Cyclotron, 89Y(p,n)89Zr
`
`228Pb/212Pb generator
`
`228Th/212Pb generator
`
`224Ra/212Pb generator
`
`a (100%)
`
`a, 5600–5830 (6)
`
`226Ra(p,2n)225Ac
`n-Capture of 232Th - 233U - 225Ac
`
`ion has different chemical demands,
`Each radiometal
`including ligand donor atom preferences (e.g. N, O, S, hard/
`soft), coordination number, and coordination geometry; how-
`ever, there are many key design considerations that can be
`applied universally.26 Ligand synthesis should be relatively
`simple and avoid stereoisomers and non-enantio/diastereospecific
`reactions, modularity should be incorporated in BFC synthesis
`as much as possible to allow for incorporation of different
`bioconjugation handles, and modular synthesis also should
`allow for tuning of denticity and physical properties by changing
`the polarity and charge of the chelate (the degree of polarity can
`
`be assessed by octanol–water partition coefficients (log P)), so
`that biodistribution properties can be adjusted.
`The intention of this article is to act as a guide for researchers
`to find the optimal match of chelator with radiometal for radio-
`pharmaceutical applications. The methods by which chelators
`are evaluated with different radiometals will be discussed, and
`the current ‘‘gold standard’’ chelators for each relevant radio-
`metal ion will be identified. A large number of review articles
`with broader scopes have been written that discuss the applica-
`tion of radiometals in radiopharmaceuticals, covering topics
`ranging from chelators, coordination chemistry, synthetic
`
`262 | Chem. Soc. Rev., 2014, 43, 260--290
`
`This journal is © The Royal Society of Chemistry 2014
`
`Published on 30 October 2013. Downloaded on 1/27/2025 6:39:21 PM.
`
`View Article Online
`
`Petitioner GE Healthcare – Ex. 1013, p. 262
`
`

`

`Review Article
`
`Chem Soc Rev
`
`chemistry of technetium, and the types of ligands used with it, differ
`drastically from those for the radiometals discussed here.26,35,52
`
`2.1 Macrocyclic versus acyclic chelators
`
`When designing new chelators, a historical glance at previous
`work reveals that macrocycles are generally more kinetically
`inert than acyclic chelators, even if their thermodynamic stabilities
`have been determined to be very similar.53–57 Macrocyclic chelators
`require minimal physical manipulation during metal
`ion
`coordination, as they possess inherently constrained geometries
`and partially pre-organized metal ion binding sites, thereby
`decreasing the entropic loss experienced upon metal ion coordi-
`nation.58 To contrast this, acyclic chelators must undergo a more
`drastic change in physical orientation and geometry in solution
`in order to arrange donor atoms to coordinate with a metal ion,
`and subsequently they suffer a more significant decrease in
`entropy than do macrocycles (thermodynamically unfavorable).
`The thermodynamic driving force towards complex formation is
`therefore greater for macrocycles in general, a phenomenon
`referred to as the macrocycle effect.58 A crucial property where
`most acyclic chelators excel and most macrocycles suffer is in the
`coordination kinetics and radiolabeling efficiency. The ability to
`quantitatively radiolabel/coordinate a radiometal in less than
`15 minutes at room temperature is a common property of acyclic
`chelators, whereas macrocycles often require heating to 60–95 1C
`for extended times (30–90 minutes).59–61 Fast room temperature
`radiolabeling becomes a crucial property when working with
`BFC-conjugates of heat sensitive molecules such as antibodies
`and their derivatives, or when working with short half-life
`isotopes such as 68Ga, 212/213Bi, 44Sc, and 62Cu.
`
`2.2 Matching chelators with radiometals – how are chelators
`evaluated?
`
`When a new chelator is synthesized for the purpose of radio-
`metal ion sequestration, or an old chelator is repurposed for
`use with a new radiometal ion, initial screening experiments
`are usually done by simple radiolabeling to determine a number
`of factors: whether the chelator can bind the radiometal ion and
`effectively radiolabel in high yields (quantitative is best), what
`temperature is required (ambient temperature is best), and what
`reaction time is required (faster is better). Short half-life isotopes
`
`Fig. 1 Examples of bioconjugation reactions:
`(A) standard peptide
`coupling reaction between a carboxylic acid and a primary amine with
`a coupling reagent;
`(B and C) peptide coupling reactions between
`activated esters of
`tetrafluorophenyl
`(TFP) or N-hydroxysuccinimide
`(NHS) and a primary amine; (D) thiourea bond formation between an
`isothiocyanate and a primary amine; (E) thioether bond formation between
`a maleimide and thiol; (F) standard Cu(I) catalyzed Huisgen 1,3-dipolar
`cycloaddition (‘‘click’’ reaction) between an azide and an alkyne; and
`(G) strain-promoted Diels–Alder ‘‘click’’ reaction between a tetrazine and
`transcyclooctene.
`
`methodologies, radiometal production, radiochemistry, bio-
`conjugation strategies, automation, molecular targeting groups
`(vectors), imaging modalities, and therapeutics.3,13,15,17,18,26–51
`Technetium chemistry is not covered here, because the
`
`Fig. 2 Illustration of an archetypal radiometal-based radiopharmaceutical agent containing a bifunctional chelator (BFC) conjugated to a targeting
`vector (e.g. antibody, peptide, nanoparticle) using a variety of conjugation methods (e.g. isothiocyanate–amine coupling, peptide coupling, maleimide–
`thiol coupling, activated ester amide coupling, click-coupling) and then radiolabeled with a radiometal ion (e.g. 111In3+/177Lu3+/86/90Y3+).
`
`This journal is © The Royal Society of Chemistry 2014
`
`Chem. Soc. Rev., 2014, 43, 260--290 | 263
`
`Published on 30 October 2013. Downloaded on 1/27/2025 6:39:21 PM.
`
`View Article Online
`
`Petitioner GE Healthcare – Ex. 1013, p. 263
`
`

`

`Chem Soc Rev
`
`Review Article
`
`like 68Ga are ideally matched with chelators that can radiolabel
`rapidly (fast radiolabeling kinetics). Longer half-life isotopes
`such as 111In and 177Lu allow for extended reaction times, but
`even if the half-life allows for long reaction times, completing the
`radiolabeling portion of radiopharmaceutical preparation is
`most convenient if finished in less than 10 or 15 minutes. As
`previously mentioned, room temperature radiolabeling is crucial
`for sensitive antibody vectors, which are degraded at elevated
`temperatures. DOTA inconveniently requires elevated tempera-
`tures for radiolabeling with essentially all radiometals (e.g. 44Sc,
`111In, 177Lu, 86/90Y, 225Ac) but its abundant application in radio-
`chemistry for decades, its exceptional in vivo stability, and the
`commercial availability of many different bifunctional DOTA
`derivatives and vector conjugates means that it is likely the most
`commonly used chelator to this day. Moving forward to the
`design and testing of new chelators, fast room temperature
`radiolabeling kinetics should be a priority; however, fast kinetics
`of radiometal
`incorporation (on-rate) and consequently low
`energetic barriers to radiometal–chelate complexation can also
`mean fast radiometal decorporation (off-rates) and low energetic
`barriers to radiometal release (Fig. 3). An arduous balancing
`act is required to obtain the best set of chelate properties for
`each application and radiometal ion, requiring the study and
`availability of a broad selection of different chelators with a
`variety of properties from which to choose.
`
`When a chelator is identified through early screening to
`possess radiolabeling properties that are suitable for use with a
`particular radiometal
`ion,
`it must then be experimentally
`determined to be highly stable and inert. Further experiments
`are performed with the specific radiometal–chelate complex
`under conditions relevant to in vivo translation to judge its
`potential as the core component of a radiometal-based radio-
`pharmaceutical. The result of radiometal loss from a radio-
`pharmaceutical in vivo is the non-targeted distribution of the
`‘‘free’’ radiometal ion in the body, and its exact fate and
`distribution in the body depends on the properties and biolo-
`gical behavior of the specific radiometal ion in question (Fig. 3).
`For example, 89Zr and 68Ga are known to accumulate in the
`bone when released from a BFC, where 64Cu is known to
`accumulate in the liver. The fate of these radiometal ions can
`be tracked using PET/SPECT imaging in the living animal, and/
`or biodistribution experiments where animals are euthanized
`at predetermined time points, their organs harvested, and the
`distribution of radioactivity measured and calculated for the
`percentage of injected dose of radioactivity per gram of tissue
`(%ID/g). Each metal ion has its own unique properties that
`must be considered when constructing a radiometal-based
`imaging/therapeutic agent, such as its aqueous hydrolysis
`chemistry, redox chemistry, and affinity for native biological
`chelators.
`
`Fig. 3 Cartoon depiction of metal ion coordination kinetics, enhanced off-rate kinetics in vivo (extremely dilute conditions), and possible routes of
`radiometal ion loss in vivo (solid-state structures of ferritin H-chain homopolymer PDB file 1FHA, ceruloplasmin PDB file 2J5W, and apo-transferrin PDB
`file 2HAV shown).
`
`264 | Chem. Soc. Rev., 2014, 43, 260--290
`
`This journal is © The Royal Society of Chemistry 2014
`
`Published on 30 October 2013. Downloaded on 1/27/2025 6:39:21 PM.
`
`View Article Online
`
`Petitioner GE Healthcare – Ex. 1013, p. 264
`
`

`

`Review Article
`
`2.3 Thermodynamic stability
`
`When evaluating and selecting a chelator to match with a
`specific radiometal
`ion for use in a radiopharmaceutical,
`kinetic inertness in vivo is ultimately the most crucial consid-
`eration, even beyond that of the absolute thermodynamic
`stability of the metal–chelate complex. Thermodynamic stability/
`formation constants (KML = [ML]/[M][L]) are usually experimentally
`determined by potentiometric and/or spectrophotometric titrations,
`but evaluating kinetic inertness under conditions relevant to in vivo
`applications can be much more problematic. Thermodynamic
`stability constants can be a useful metric for preliminary
`comparisons of various chelators with a particular metal ion,
`but they do not predict in vivo stability with any level of
`competence.62,63 A thermodynamic parameter that provides
`more biologically relevant information than KML values is the
`pM value (log[M]Free).64–67 The pM value is the negative log of
`the concentration of free metal ion uncomplexed by a given
`chelator under specific conditions. The pM value is a condition-
`dependent value that is calculated from the standard thermo-
`dynamic stability constant (log KML), accounting for variables
`and conditions such as ligand basicity, metal ion hydrolysis,
`(physiological) pH, and ligand : metal ratio.
`Stability constants and pM values provide a number for the
`direction and magnitude of the equilibrium in a metal–chelate
`coordination reaction under specific conditions, but give no
`kinetic information (e.g. off-rates for dissociation).68,69 This is a
`very important factor because the rate of dissociation in vivo is
`what governs the kinetic inertness of a radiometal complex,
`regardless of thermodynamic stability, and these off-rates are
`greatly influenced by the high dilution encountered in vivo
`when a tiny quantity of radiopharmaceutical (micro- or nano-
`grams) is diluted into the blood pool (Fig. 3). Even more
`complicating is the abundant presence in the body of many
`strong native biological chelators and competing ions that can
`transchelate radiometals from BFC-conjugates. These are often
`present in higher concentrations than is the radiopharmaceutical,
`and include transport proteins such as transferrin, ceruloplasmin,
`and metallothionein, storage proteins like ferritin, and metal
`containing enzymes such as superoxide dismutase (Fig. 3). This
`wide range of complicating factors means that a single in vitro assay
`is typically not accurate in predicting in vivo stability.
`
`2.4 Kinetics – acid dissociation and competitive radiolabeling
`
`Acid dissociation experiments can be used to measure and
`assess the relative kinetic inertness of a metal complex to acidic
`conditions. Most complexes are found to de-ligate fairly quickly
`below pH 2.0,70 and experiments evaluating the rate of dis-
`sociation at a constant pH (e.g. 2.0) have been performed to
`compare and evaluate chelators with a particular metal/radio-
`ion.55,70,71
`metal
`In these acid dissociation experiments,
`decomplexation can be observed over time with techniques
`such as high-performance liquid chromatography (HPLC), thin-
`layer chromatography (TLC), and nuclear magnetic resonance
`(NMR) for diamagnetic metal complexes. With the exception of
`copper chelates,68,72 acid dissociation experiments are not
`
`Chem Soc Rev
`
`commonly performed as they do not provide an accurate
`prediction of in vivo kinetic inertness, because low pH is not
`encountered in the blood or most organs (except the stomach)
`and radiometal dissociation typically occurs via transchelation to
`serum proteins and enzymes, and is not acid-mediated.69,73–77
`Competitive radiolabeling experiments can be performed by
`adding to a radiolabeling mixture an excess of non-radioactive
`ions, such as Na+, K+, Ca2+, Mg2+, Cu2+, Zn2+, or Fe3+, followed
`by addition of a chelator to evaluate the impact of these
`competing ions on radiolabeling yields.55,78–81 These experi-
`ments can reveal the radiolabeling selectivity of a chelator for a
`specific radiometal ion in the presence of other biologically
`relevant competing ions. Alternatively,
`if
`the radiometal
`complex is preformed under standard metal-free radiolabeling
`conditions, and is then added to a mixture containing these
`competing ions, stability to transchelation can be assessed.
`Because chelate-based radiopharmaceuticals are typically radio-
`labeled in strictly metal-free conditions (deionized ultrapure
`water, often passed through a metal-scavenging chelex resin),
`these experiments may not appear completely relevant for pre-
`dicting in vivo stability and utility. Depending on the method of
`radiometal production and purification, the specific activity of
`radiometal ions can vary greatly, as can the concentrations of
`impurity metals ions.5 Radiometals are used in very small
`quantities and under extremely dilute conditions; therefore,
`any impurity metal ions present (even if only a few nanomoles)
`may actually be in excess of the radiometal ion concentration,
`and competitive binding could be problematic.5,82 The presence
`and quantity of metal ion impurities in radiometal mixtures
`depends on the source of the radiometal ion, method of produc-
`tion, and purification.5,82
`
`2.5 In vitro and in vivo stability
`
`Experiments that are more pertinent to in vivo translation are
`metal-exchange competitions with biologically relevant mixtures,
`including blood serum, apo-transferrin, superoxide dismutase,
`and hydroxyapatite (bone).18,83–88 By incubating radiometallated
`chelators with these different competition mixtures,
`the
`quantity of radiometal that is transchelated from chelator to
`serum proteins/enzymes can be evaluated over time using
`size exclusion HPLC, iTLC, or disposable PD10 size exclusion
`columns.18,83–88 These experiments provide a directly relevant
`measure of stability and kinetic inertness by competition with
`the most likely transchelation culprits in vivo; additionally, in
`these in vitro assays the concentrations of these biological
`reagents can often be elevated above normal physiological
`levels to provide a more stringent challenge.
`Ultimately the most relevant and practical test of radio-
`pharmaceutical stability is in vivo. Biodistribution studies in
`healthy mice can be performed on ‘‘naked’’ radiometal–chelate
`complexes (no conjugated vectors) to assess clearance and
`uptake profiles and ensure no abnormal organ distribution or
`critical instability occurs. If a radiometal–chelate complex is
`very stable in vivo, the complex is typically cleared quickly from
`the animal through the kidneys/bladder/urine or digestive
`system/liver/feces depending on polarity and metabolism.89
`
`This journal is © The Royal Society of Chemistry 2014
`
`Chem. Soc. Rev., 2014, 43, 260--290 | 265
`
`Published on 30 October 2013. Downloaded on 1/27/2025 6:39:21 PM.
`
`View Article Online
`
`Petitioner GE Healthcare – Ex. 1013, p. 265
`
`

`

`Chem Soc Rev
`
`Review Article
`
`Unstable complexes often demonstrate persistent uptake in
`organs and tissues where the non-bound radiometal is known
`to associate (e.g. Zr4+ and lanthanides in the bone, Cu2+ in the
`liver).89 The major drawback to this type of experiment with
`‘‘naked’’ chelate complexes is that highly polar and charged
`radiometal–chelate complexes are typically cleared very quickly
`from the body, and therefore do not persist
`in vivo for
`long enough to encounter any significant challenge to their
`structural integrity.84,85 Conversely, highly lipophilic complexes
`tend to accumulate in the liver and digestive tract, regardless of
`stability.84,85
`To evaluate properly the stability and kinetic inertness of a
`chelate in vivo, a suitable vector must be attached (e.g. peptide,
`antibody, nanoparticle) so that the radiometal–chelate complex
`is made to persist in blood circulation for a substantial period
`of time, and can be monitored over several hours or days
`(depending on isotope half-life and the subsequent imaging/
`therapy window).90 An additional concern with in vivo experi-
`ments is the significant normal variance between animals
`that introduces error; for example, 10 mice of the same sex
`and breed, procured from the same supplier, will often show
`significant differences in biodistribution of the same radio-
`pharmaceutical. Additionally, the specific experimental techniques
`and methods (e.g. radiopharmaceutical preparation,
`injection,
`animal dissection, organ counting) used for these biodistribution
`experiments can cause large variability between data sets. Variables
`such as specific activity of the source radiometal, specific activity of
`the radiolabeled agent, radiolabeling temperature, purity of the
`final radiopharmaceutical preparation, injected mass of radiophar-
`maceutical, and injected volume can make large differences in
`biodistribution profiles. The same radiopharmaceutical used at
`different institutions may offer drastically different tumor uptake
`and organ distribution values, if the above variables are not tightly
`controlled, and even then differences between animals and
`between experimental techniques can introduce large variances.
`For these reasons it is crucial to include internal control experi-
`ments for every study; for example, evaluation of a new chelator
`in vivo should be done in parallel with an existing and established
`‘‘gold standard’’ chelator so that a direct comparison can be made
`because comparison to previously performed studies (even in the
`exact same animal model and/or cell line) is often not reliable due
`to the above mentioned complications.79,83,85,91
`
`3. A selection of chelators and their
`most suitable radiometal companions
`
`A survey of radiometal chelators has been undertaken, with the
`most relevant and promising examples being discussed. The
`most suitable chelate–radiometal matches have been identi-
`fied, and detailed tables included, with the aim of providing a
`quick reference for those looking to find the optimal match
`between chelator and radiometal. A color-coding scheme has
`been used, and justification for the assignment of good (green),
`fair (orange), and poor (red) matches between chelators
`and radiometals can be found in the text and the provided
`
`references. The color code indicates the authors’ opinions on
`the general suitability of a chelator for use with a specific
`radiometal, accounting for
`factors such as radiolabeling
`conditions and in vitro/in vivo stability. An assignment of green
`may suggest that either a chelator is currently the ‘‘gold
`standard’’ for use with a given radiometal, or that early work
`with a new chelator looks very promising and in vivo studies
`have shown that it works comparably to the current ‘‘gold
`standards’’. Assignment of orange to a chelator–radiometal
`pair may be made if the combination has been used in vivo
`successfully, but perhaps in recent years has been surpassed by
`a new and superior chelator and is no longer the best choice.
`Also, an assignment of orange could mean that a new chelator–
`radiometal ion pair looks very promising, but perhaps a bifunc-
`tional derivative has yet to be synthesized, or only preliminary
`in vivo work has been done and more study is required (e.g. no
`study of bioconjugat