Influence of Metallic Cations on DOTA Labeling with 90Y and 177Lu PDF

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/257432093 Influence of metallic cations on the labelling reaction of DOTA with 90- Yttrium and 177-Lutetium Article in Nuclear Medicine and Biology · August 2010 DOI: 10.1016/j.nucmedbio.20...

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/257432093 Influence of metallic cations on the labelling reaction of DOTA with 90- Yttrium and 177-Lutetium Article in Nuclear Medicine and Biology · August 2010 DOI: 10.1016/j.nucmedbio.2010.04.096 CITATIONS READS 0 2,169 6 authors, including: Mattia Asti Michele Iori Azienda Unità Sanitaria Locale Reggio Emilia IRCCS - Santa Maria Nuova Reggio Emilia 77 PUBLICATIONS 1,081 CITATIONS 31 PUBLICATIONS 417 CITATIONS SEE PROFILE SEE PROFILE Claudio Guidotti Annibale Versari Azienda Ospedaliera Santa Maria Nuova di Reggio Emilia AUSL-IRCCS of Reggio Emilia 7 PUBLICATIONS 95 CITATIONS 304 PUBLICATIONS 7,723 CITATIONS SEE PROFILE SEE PROFILE All content following this page was uploaded by Mattia Asti on 26 October 2017. The user has requested enhancement of the downloaded file.  Available online at www.sciencedirect.com Nuclear Medicine and Biology xx (2011) xxx – xxx www.elsevier.com/locate/nucmedbio Influence of cations on the complexation yield of DOTATATE with yttrium and lutetium: a perspective study for enhancing the 90 Y and 177 Lu labeling conditions Mattia Asti a,⁎, Matteo Tegoni b , Daniela Farioli a , Michele Iori a , Claudio Guidotti a , Cathy S. Cutler c , Pat Mayer d , Annibale Versari a , Diana Salvo a a Nuclear Medicine Department, Santa Maria Nuova Hospital, Reggio Emilia, Italy b Department of General and Inorganic, Analytical and Physical Chemistry, University of Parma, Parma, Italy c University of Missouri, Research Reactor Center, Columbia, Missouri, US d PerkinElmer, Waltham, Massachusetts, US Received 11 April 2011; received in revised form 14 September 2011; accepted 22 October 2011 Abstract The DOTA macrocyclic ligand can form stable complexes with many cations besides yttrium and lutetium. For this reason, the presence of competing cationic metals in yttrium-90 and lutetium-177 chloride solutions can dramatically influence the radiolabeling yield. The aim of this study was to evaluate the coordination yield of yttrium- and lutetium-DOTATATE complexes when the reaction is performed in the presence of varying amounts of competing cationic impurities. In the first set of experiments, the preparation of the samples was performed by using natural yttrium and lutetium (20.4 nmol). The molar ratio between DOTATATE and these metals was 1 to 1. Metal competitors (Pb 2+, Zn 2+, Cu 2+, Fe 3+, Al 3+, Ni 2+, Co 2+, Cr 3+) were added separately to obtain samples with varying molar ratio with respect to yttrium or lutetium (0.1, 0.5, 1, 2 and 10). The final solutions were analyzed through ultra high-performance liquid chromatography with an UV detector. In the second set of experiments, an amount of 90Y or 177Lu chloride (6 MBq corresponding to 3.3 and 45 pmol, respectively) was added to the samples, and a radio-thin layer chromatography analysis was carried out. The coordination of Y 3+ and Lu 3+ was dramatically influenced by low levels of Zn 2+, Cu 2+ and Co 2+. Pb 2+ and Ni 2+ were also shown to be strong competitors at higher concentrations. Fe 3+ was expected to be a strong competitor, but the effect on the incorporation was only partly dependent on its concentration. Al 3+ and Cr 3+ did not compete with Y 3+ and Lu 3+ in the formation of DOTATATE complexes. © 2011 Elsevier Inc. All rights reserved. 90 177 Keywords: Y-DOTATATE; Lu-DOTATATE; Cationic metals; PRRT 1. Introduction clides to tumors expressing somatostatin receptors. For peptide receptor radionuclide therapy, radiolabeling specific The modified version of the coordinating agent 1,4,7,10- activity (RSA) — i.e., radionuclide–ligand molar ratio — tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) C- should be as high as possible to minimize receptor functionalized is used as a bifunctional ligand for the occupancy of unlabeled conjugates, thus preventing dose radiolabeling of biological analogues such as antibodies and adjustments. Thus, optimization of radiolabeling efficiency peptides. In particular, DOTA-conjugated somatostatin would affect peptide receptor radionuclide therapy efficacy. analogues, such as DOTA 0-Tyr 3-octreotide and DOTA 0- Because of the slow kinetics of the coordination reaction Tyr 3 -octreotate (DOTATATE), are clinically used for [2,3] to achieve the maximum chelation efficiency, labeling delivering yttrium-90 and lutetium-177 therapeutic radionu- procedures of clinical-grade radiopeptide generally are performed with a molar excess of ligand. However, the specific activity in the commercially available radionuclide ⁎ Corresponding author. Tel.: +39 0522 296766; fax: +39 0522 296153. solutions — i.e., the ratio of the radionuclide activity divided E-mail address: [email protected] (M. Asti). by the mass of the stable carrier/s — as well as the presence 0969-8051/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.nucmedbio.2011.10.015 2 M. Asti et al. / Nuclear Medicine and Biology xx (2011) xxx–xxx of metal contaminants should be considered in order to from Sigma-Aldrich (Milan, Italy). All metal chlorides were achieve the highest RSA and the total completion of the the highest purity available (generally N99.9%), and the coordination reaction. presence of other metals in the purchased lots was less than The DOTA moiety is not selective for therapeutic 0.05% wt/wt, as reported on the certificates of analysis. radionuclides such as 90Y or 177Lu since it may form Metal free hydrochloric acid (0.05 M) was purchased from thermodynamic and kinetically stable complexes with many Carlo Erba (Milan, Italy), and milliQ water (resistivity other cations [4,5], so the presence of cationic impurities in 18.2 MΩ·cm) was used for every preparation listed in this 90 Y or 177Lu chloride solutions can dramatically influence paper. DOTATATE peptide, with N99.0% purity, was the radiolabeling yield, decreasing radiochemical purity. obtained from ABX (Radeberg, Germany). As reported on In fact, cation competition effects on the radiolabeling of the certificate of analysis, the amount of Zn and Fe in this lot DOTA with 90Y and 177 Lu have been described [2,3], but a was 4·10 −4% and 1.4·10 −3% wt/wt, respectively, while Cu, detailed quantification of the minimum amounts of contam- Ni and Pb amounts were under the limit of detection inants assuring the completion of the coordination reaction (b2·10 −4%). Yttrium-90 chloride (18,500 GBq/mg specific has not been reported. Moreover, while the interference of activity) and lutetium-177 chloride (740 GBq/mg specific some metallic contaminants (Hg 2+, Ga 3+, Y 3+, Cu 2+ and activity) solutions (62 MBq in 1 ml) were supplied by Perkin Cd 2+) was specifically examined for 177 Lu-DOTATATE Elmer (Boston, MA, USA). The lots were provided with a radiolabeling , a quantitative study on the competition certificate of analysis attesting that the cationic impurities with the same conditions used for a clinical-grade radio- amounts were below the limit of detection of the analytical pharmaceutical preparation has not been evaluated. To our method applied. The limits of detection were the following: knowledge, there are few quantitative studies on the Zn b0.3 μg/Ci; Al b0.8 μg/Ci; Cu b0.3 μg/Ci; Fe b0.2 μg/Ci; influence of metal contaminants on 90Y DOTA-peptide Co b0.1 μg/Ci; Ni b 0.1 μg/Ci; Pb b2.0 μg/Ci and Cr b0.4 labeling, and only the effects of Fe 2+ and Cu 2+ have been μg/Ci. All the reagents were used without further purifica- reported. Recently, a more complete study about the tion. The metal-DOTATATE complex formation was influence of chemical impurities (Ca, Cd, Cu, Fe and Zn) on assessed by ultra high-performance liquid chromatography both 90Y and 177 Lu-DOTATATE labeling was published (UHPLC) using an Acquity system with a binary solvent, , but the results are difficult to compare, and a general BEH C-18 1.7-μm (2.1×150 mm) column and autosampler behavior for the single cations is not deducible. manager modules (Waters, Milan, Italy). The instrument was In this paper, we describe a perspective study where the equipped with an Acquity TUV detector (Waters, Milan, coordination yields of both yttrium- and lutetium-DOTA- Italy). In the presence of yttrium-90 or lutetium-177 chloride TATE were evaluated when the reactions occurred in the (vide infra), the radiochemical yield was assessed by thin presence of varying amounts of different metal impurities layer chromatography (TLC) using an AR 2000 Imaging and for a peptide/radionuclide molar ratio equal to 1. The Scanner device (Bioscan, Washington, DC, USA). The metal impurities were selected from the transition metals that electrospray ionization–mass spectrometry (ESI-MS) was complex with high affinity to the DOTA ligand, namely, performed by using an LTQ Orbitrap XL instrument Fe 3+, Ni 2+,Co 2+, Cu 2+ and Zn 2+. In addition, the effect of (Thermo Scientific, Waltham, MA, USA). Cr 3+, Pb 2+ and Al 3+ was assessed. Experimental data were compared with the results obtained by theoretical calculation 2.2. Speciation calculations based on the reported thermodynamic constants available for DOTA and DO3A (1,4,7,10-tetraazacyclododecane-1,4,7- The calculation of the theoretical amounts of metal triacetic acid) at 25°C to elucidate which factors, apart from complexes present at the equilibrium was performed using metal complex stability in standard conditions, may affect the stability constants for metal:DOTA or metal:DO3A labeling yields in the presence of metal contaminants. The (DO3A=1,4,7,10-tetraazacyclododecane-1,4,7-triacetic aim of the study was to obtain a complete profile of Y- and acid) available in the literature [10,11], as summarized in Lu-DOTATATE yields in relation to the ratio between the Table 1. These constants are reported for aqueous solutions at nuclide, the metal impurity considered and the experimental 25°C. The calculations were performed for samples contain- conditions used. These results can be transferred to a clinical ing the same quantities of ligand and metal as those used for radiopharmaceutical preparation and used to optimize the UHPLC analyses (Y, Lu=20.4 nmol; DOTA or labeling conditions. DO3A=20.4 nmol; metal contaminant=2.04–204 nmol) and the same pH (4.6). The stability constants of hydroxocom- plexes and complexes with acetate (present as buffer) were 2. Materials and methods included in the calculations. The calculations were 2.1. Reagents and instrumentation performed using the Hyss2006 (Hyperquad Simulation and Speciation) software. DOTA and DO3A were chosen for Yttrium(III), lutetium(III), lead(II), aluminium(III), chro- calculation because a homogeneous set of stability constants mium(III), iron(III), nickel(II), cobalt(II), copper(II) and zinc referred to as amide-functionalized DOTA ligands (as a (II) chlorides and sodium acetate anhydrous were purchased model for DOTATATE) is not available for these cations.  M. Asti et al. / Nuclear Medicine and Biology xx (2011) xxx–xxx 3 Table 1 samples were as follows: (a) a 3 ml Schott vial was filled Log β values for the M+DOTA=[M(DOTA)] and the M+DO3A= with 500 μl of a 0.05 M HCl solution; (b) 20.4 nmol of [M(DO3A)] equilibrium natural yttrium or lutetium chloride in 0.05 M HCl was Metal Log β Log β added to each vial; (c) varying amounts of metal cations were M+DOTA=[M(DOTA)] M+DO3A=[M(DO3A)] added to the vials (from 0.1 to 10 equivalents with respect to Y 3+ 24.0 21.1 yttrium or lutetium) using proper amounts of 0.05 M HCl Lu 3+ 23.9 23.0 solutions of metal chlorides; (d) each sample was buffered to Al 3+ 17.0 NA Co 2+ 20.27 NA pH 4.6 by addition of a 0.2 M sodium acetate solution; (e) Cu 2+ 22.44 21.65 20.4 nmol of DOTATATE from a 1 mg/ml water solution Fe 3+ 29.4 NA was added to each vial; (f) each test vial was heated for Ni 2+ 20.03 NA 30 min at 90°C in a heating block. Every sample was Pb 2+ 22.69 NA analyzed by UHPLC, and the chromatograms were com- Zn 2+ 20.52 19.3 Cr 3+ NA NA pared to the reference standards obtained as described before. The percentage of Y- and Lu-DOTATATE complex NA=not available. in every sample was computed as the relative ratio of their UHPLC signal to any metal-DOTATATE complex detected 2.3. Preparation of the metals-DOTATATE used as in the solution. The integration of the UHPLC peaks was reference standards performed automatically by Empower software (Waters, Milan, Italy). All reactions were carried out in triplicate. In order to perform a UHPLC analysis for every complex and obtain the retention times of metal-DOTATATE species, 2.5. Preparation of the samples analyzed by each metal chloride was reacted with a DOTATATE solution using radio-TLC separately. The reactions were performed by addition of 20.4 nmol of metal in 0.05 M HCl metal chloride solution UHPLC is a more accurate analytical method but is not (metals=Y 3+, Lu 3+, Pb 2+, Zn 2+, Cu 2+, Fe 3+, Al 3+, Ni 2+, routinely available to most clinical laboratories, so radio- Co 2+, Cr 3+) to 0.5 ml of a 0.05-M HCl solution. The samples TLC which is commonly used was performed to determine were buffered to pH 4.6 by using a 0.2 M sodium acetate the radiolabeling yield, and the results were compared to solution and then adding 20.4 nmol of DOTATATE from a those obtained by UPLC. A volume of 0.1 ml containing 1 mg/ml water solution to each sample. The final volume about 6 MBq of 90Y or 177 Lu chloride solution in 0.05 M was constant for each sample. The resultant mixtures were HCl with a maximal amount of additional contaminants of heated for 30 min at 90°C in a heating block. Every metal- 1.6 pmol for Pb 2+ and 4.9 pmol for Al 3+ (calculated from DOTATATE complex solution was injected onto the data reported on the certificate of analysis provided by the UHPLC system, and the chromatograms were used as supplier) was added to each sample between steps (b) and reference standards for the following set of experiments. The (c) of the procedure described above. These concentrations UHPLC analyses were performed in isocratic conditions were considered negligible compared to the metals amounts with a 27% acetonitrile and 73% 0.1% vol/vol TFA water added to the sample in step (c). Yttrium-90 and lutetium- solution as mobile phase at a flow rate of 0.35 ml/min. The 177 mass was 3.3 pmol and 45 pmol, respectively, while wavelength of the UV detector was set to 220 nm, and the 20.4 nmol of nonradioactive nuclides was used in the column temperature was fixed to 30°C. The formation and sample preparations. Radioactivity of the radionuclides was identity of the metal-DOTATATE complexes were con- sufficient to carry out evaluation of radiochemical purity, firmed through their m/z ratio and isotopic patterns obtained while the chemical mass of each radionuclide was three by ESI-MS analysis. A 100-ppm DOTATATE solution was orders of magnitude lower than that of the nonradioactive employed to determine the retention time of the free ligand nuclides; therefore, it was not taken into account when after each set of runs. establishing metal:peptide molar ratios. 2.4. Preparation of the samples analyzed by using UHPLC The radiochemical yield of 90Y- and 177Lu-DOTATATE was determined by radio-TLC with a 97% 0.1 M sodium Sample preparation was carried out in the same condition citrate and 3% 1 M HCl solution as mobile phase (pH 4.5) as for a clinical-grade radiolabeling procedure, as reported and TLC silica gel 60 RP-18 F254 plates (Merck, Milan, by Grassi et al.. Natural yttrium and lutetium chloride Italy) as stationary phase, respectively. When these condi- solutions in place of yttrium-90 and lutetium-177 solutions tions were applied, 90Y-DOTATATE or 177Lu-DOTATATE and a nuclide:DOTATATE molar ratio of 1 were used in exhibited Rf=0.3, while unlabeled 90Y 3+ or 177Lu 3+ moved every sample. The pH of the reactions was buffered around with an Rf=0.7. The radiochemical yield of the reaction was 4.6 by using a 0.2 M sodium acetate solution. Each cationic computed for all samples as the relative ratio between the competitor (Pb 2+, Zn 2+, Cu 2+, Fe 3+, Al 3+, Ni 2+, Co 2+, Cr 3+) radio-TLC signal belonging to 90Y- or 177Lu-DOTATATE was added separately in five different molar ratios (0.1, 0.5, to the sum of all signals present in the chromatograms. Every 1, 2 and 10) with respect to natural yttrium or lutetium. All reaction was carried out in triplicate. 4 M. Asti et al. / Nuclear Medicine and Biology xx (2011) xxx–xxx 2.6. Data treatment findings that report an excess of ligand (twofold and threefold chelator:metal molar ratio for Lu-177 and Y-90, Data obtained by using the two analytic methods respectively) is necessary to obtain a N90% radiochemical (UHPLC and radio-TLC) were compared using linear yield [2,3]. This difference may be related to the higher regression, and the agreement between the data from the specific activity of Lu-177 (N740 GBq/mg) and the higher two techniques was verified through the linear correlation radionuclidic purity of Y-90 ( 90Sr/ 90Y ratiob10 −7%) used in coefficient (R 2). Results presented in this study are expressed this study with respect to the reported ones and is in as means. Standard deviations were computed but were not agreement with data recently reported by Urbano et al. , reported in the figures. Data were treated by using an SPSS where a DOTA-conjugated biotin analogue was labeled with v.18 software (IBM, Chicago, IL, USA). Y-90 or Lu-177 at a very high RSA. 3.2. Competition of cationic metals in the 3. Results and discussion coordination reactions 3.1. Preparation of the reference standards The experimental results for the Y/Lu-DOTATATE In Fig. 1, an overlapping of the metal-DOTATATE complex formation when the reaction was performed in chromatograms and the relative retention times is depicted. the presence of varying amounts of metal contaminants As shown, a good separation among the signals belonging (M=Pb 2+, Zn 2+, Cu 2+, Fe 3+, Al 3+, Ni 2+, Co 2+) are reported to the metal-DOTATATE and Y- or Lu-DOTATATE in Fig. 2 (A,D), while the theoretical amounts expected from signal was achieved, allowing a reliable determination of the distribution studies with DOTA or DO3A are summa- the relative percentage when the competition study was rized in Fig. 2 (B,C,E,F). Plots B and E were drawn using performed. Single chromatograms showed the complete Y 3+/Lu 3+:DOTA stability constants; plots C and F were disappearance of the signal related to free DOTATATE drawn using Y 3+/Lu 3+:DO3A stability constants. (b 3%) and the formation of new signals at different Because of the lack of available data on the thermody- retention times, indicating the completion of the coordina- namic stability of metal/DOTATATE complexes and the tion reaction for all the cations, with the exception of Al 3+ corresponding formation constants; we performed two and Cr 3+. These cations did not form stable complexes with different sets of calculations: (a) estimation of the theoretic DOTATATE, and no new signals could be observed in amounts of Y 3+ or Lu 3+/DOTA in the presence of the their chromatograms. contaminants and (b) estimation for Y 3+ or Lu 3+/DO3A Radiochemical yield was always greater than 97% even if systems, using for the contaminants the log β values a molar ratio equal to 1 was used. This differs from published with DOTA. Fig. 1. UHPLC chromatograms and retention times of the metals-DOTATATE complexes.  M. Asti et al. / Nuclear Medicine and Biology xx (2011) xxx–xxx 5 Fig. 2. Experimental coordination yield of Lu/Y-DOTATATE (A, D) and theoretic coordination yield of Lu/Y-DOTA (B, E) or Lu/Y-DO3A (C, F) in the presence of metal impurities in a molar ratio from 0.1 to 1 in respect to Y 3+ or Lu 3+. The molar ratio between Y 3+ or Lu 3+ and DOTATATE was equal to 1. Data calculated using the stability constants for metal contaminants with DO3A are marked with an asterisk and indicated as open symbols. All other data were calculated using the metal:DOTA stability constants. In fact, it has been clearly established in the literature that DOTA and DO3A, as in DOTATATE the fourth pendant the functionalization of one carboxylate group reduces the arm is present and the interaction of the cation with the amide stability of the metal/macrocyclic chelate complexes and, oxygen is still possible. Unfortunately, the log β values for therefore, DOTA complexes are more stable than those the metal contaminants considered herein with DO3A are not where one carboxylic group is either removed or functiona- available, with the exception of Cu 2+ and Zn 2+. So, the plots lized as an amide [10,11,15,16]. However, while no data are for metal contaminants shown in Fig. 2 (C,F) were drawn available in terms of the complexation constant of amide using M:DOTA stability constants except for Cu 2+ and derivatives of DOTA with Lu 3+ and Y 3+, these data are Zn 2+, where yields were calculated with both M:DOTA and available for the DO3A ligand corresponding to DOTA M:DO3A. Data for Cr 3+ are not reported as the radionuclide lacking one of the four carboxylic pendant arms. DO3A complexation yield, for every Cr 3+/radionuclide molar ratio complexes with Lu 3+ or Y 3+ are less stable than those with was close to 100% and its stability constants with the DOTA, as shown by the log β values which drop from 24 to macrocycles has not been reported. Thus, the second set of 21.1 for Y 3+ and from 23.9 to 23 for Lu 3+, respectively. calculations contains the approximation from DOTA to The values of the stability constants of Y- and Lu- DO3A, which more correctly represents the stability DOTATATE complexes are expected to be within those of constants of Y 3+ or Lu 3+ by the reduction of ligand denticity 6 M. Asti et al. / Nuclear Medicine and Biology xx (2011) xxx–xxx than those of the other metals. This is partially due to the (either for DOTA or DO3A) and that obtained experimen- smaller size of the contaminants in relation to Y 3+ and Lu 3+ tally for radionuclide-DOTATATE. In fact, the experimen- and to a coordination number close to 6 with DOTA in the tal data demonstrate a dramatically reduced formation yield solid state. Therefore, the exclusion of one carboxylate for both Y 3+ and Lu 3+ by Zn 2+, Cu 2+ and Co 2+ (b70% at group has major impact on the stability of the complexes 0.1 molar ratio for both yttrium and lutetium). In particular, with 8- or 9-coordination fold, such as Ln 3+ or Y 3+, as the behavior of Zn 2+, a very effective competitor for compared to ones formed with the contaminants. This is not DOTATATE even at 0.1 molar ratio, is not predictable strictly the case for Cu 2+ and Zn 2+, in which the log β values based on its stability constants with both DOTA and DO3A, decrease from 22.44 to 21.65 and from 20.52 to 19.3 if which would suggest a negligible effect on the radionuclide: DOTA or DO3A are considered, respectively. Nevertheless, ligand formation yield. A relative consistency between the calculated amounts of complexed Y 3+ or Lu 3+ with experimental and computed data was found for Pb 2+ and DOTA and DO3A using this set of data should represent the Ni 2+, which were shown to be strong competitors against upper and lower values for the Y 3+ or Lu 3+/DOTATATE Y 3+ but at higher concentrations (b70% at 0.5 molar ratio system, respectively. These diagrams illustrate the fact that for lutetium and b50% at 0.5 molar ratio for yttrium). Fe 3+ Y 3+ and Lu 3+ complexes with DO3A are more sensitive than did show a peculiar behavior and can be used as a relevant those with DOTA to the presence of contaminants, leading to example to demonstrate how the thermodynamic parameters a reduced Y 3+ or Lu 3+:ligand formation yield as expected are not sufficient to explain the observed experimental with the lower stability. In all diagrams, the dependence of behavior. In fact, since the stability constant of the Fe(III): Y- and Lu-DOTATATE formation (DOTA for speciation DOTA is six orders of magnitude higher than those of Lu calculations) on cationic concentrations is shown between and Y, for Fe:DOTAb1, a complete coordination of Fe to 0.1 and 1 molar ratio for Pb 2+, Zn 2+, Cu 2+, Fe 3+, Ni 2+ and DOTA is expected with only the remaining part of the Co 2+. Fig. 3 demonstrates the agreement between the two ligand available for Y and Lu complexation (expected sets of experimental data obtained by using UHPLC and yields ranging from 90% to 0%). On the other hand, when radio-TLC for lutetium and for yttrium (R 2=0.986 and Fe:DOTAN1, no free DOTA is expected to be available in R 2=0.965, respectively). Thus, Fig. 2 (A,D) presents only solution for radionuclide complexation (expected the results obtained by UHPLC analysis. A representative yield=0%). Nevertheless, the experimental yield results are UHPLC chromatogram is shown in Fig. 4. higher than 50% over the entire Fe:Y(Lu) range explored. A comparison of the data presented in Fig. 2 shows a low This trend can be explained considering the low stability of correlation between the amount of radionuclide-DOTA Fe 3+ in pH 4.6 solutions where the cation may form highly computed on the basis of the reported stability constants insoluble hydrolyzed products. The reduction of Fe 3+ to Fe 2+, although unlikely, may also justify the reduced coordination capabilities, as DOTA complexes with Fe 2+ 120 are markedly less stable than those of Fe 3+ (log β of Fe 2+/ RCY radio-TLC (%) 100 DOTA=20.22 at 25°C , ca. 10 orders of magnitude Lu-DOTATATE 80 lower than that of Fe 3+). 60 R2 = 0,9869 Both Al 3+ and Cr 3+ did not compete with either Y 3+ or 3+ 40 Lu in the formation of DOTATATE complexes up to 10 molar ratios, as expected due to the low stability of the Al 177 20 (III) complex with DOTA (for the very inert Cr 3+, the 0 0 20 40 60 80 100 120 stability constant is not reported). In particular, the small size Lu-DOTATATE Coordination Yield UHPLC (%) of the Al 3+ cation may prevent it from fitting in the DOTA A cavity; therefore complexes with DOTATATE are not 120 formed. Indeed, the high inertness of Cr 3+ chloro/acquo complexes prevent the formation of DOTATATE RCY radio-TLC (%) 100 Y-DOTATATE complexes because of the extremely slow kinetics in the 80 substitution of the coordinating molecules. 60 R2 = 0,9658 In general, if cationic radii of the metal contaminants are 40 considered, a quite strong correlation was found between the 90 20 influence of the metals on the coordination yield of Y and 0 Lu-DOTATATE and the radii themselves. Fig. 5 illustrates 0 20 40 60 80 100 120 the dependence of Y and Lu-DOTATATE coordination Y-DOTATATE Coordination Yield UHPLC (%) yields in relation to the competing cation ionic radius when B the molar ratio was 1 to 1. The ionic radii are all six- Fig. 3. Linear correlation between the two sets of labeling efficiency data coordinate except for Fe 3+, where seven-coordination was obtained by using UHPLC and radio-TLC for Lu-DOTATATE (A) and considered [17,19,20]; thus these results can be used to Y-DOTATATE (B). correlate the tendency of forming more stable metal/  M. Asti et al. / Nuclear Medicine and Biology xx (2011) xxx–xxx 7 0,30 0,28 1: Y-DOTATATE 2: Ni-DOTATATE 0,26 1 0,24 Molar Ratio Ni / Y = 1 0,22 Molar Ratio Ni / Y = 0.5 Molar Ratio Ni / Y = 0.1 0,20 0,18 0,16 0,14 2 0,12 0,10 0,08 0,06 0,04 0,02 0,00 -0,02 1,30 1,40 1,50 1,60 1,70 1,80 1,90 2,00 2,10 2,20 2,30 2,40 2,50 Minutes. 0.1364 AU Minutes Fig. 4. Typical UHPLC chromatogram obtained when Y 3+ was reacted with DOTATATE in the presence of Ni 2+ in a molar ratio ranging from 0.1 to 1. Analogous chromatograms were obtained for Lu 3+ or Y 3+ reactions with DOTATATE in the presence of the other cations. DOTATATE hexacoordinated complexes with the metal number of Pb 2+ with DOTA ligands have been reported and radius itself. As shown in the figure, results obtained in the a six-coordination number for Pb-DOTA is difficult to competition for both Y 3+ and Lu 3+ coordination are predict, as higher coordination numbers are expected in comparable, and the maximal interference was found for analogy with Y 3+ and Lu 3+ that show eight- and nine- metals with an ionic radius ranging between 0.730 and 0.745 coordination, respectively. Dimension is not the only Å (Cu 2+, Zn 2+ and Co 2+). From this direct comparison, parameter influencing the cations' affinity for DOTATATE, we can conclude that the reaction forming DOTATATE but other variables such as coordination geometry or chloro/ hexacoordinated complexes is favored for contaminants with aquo ions complex lability should also be considered. an ionic radius around 0.74 Å. The Pb 2+ behavior was not The data demonstrate that the stability constants for considered in this analysis, as no data about the coordination metal-DOTA complexes (DOTA used as the model for other Fig. 5. Correlation between Lu-DOTATATE (A) or Y-DOTATATE (B) coordination yields and the cationic radius of the metals tested. The data were taken from the UHPLC results when the molar ratio was 1 to 1. 8 M. Asti et al. / Nuclear Medicine and Biology xx (2011) xxx–xxx DOTA-based ligands) may not be the only parameter for the cations are competing for complex formation in the same evaluation of the radiometallation yields under our employed solution. In these cases, the kinetic constants of the reactions experimental conditions. Furthermore, DO3A is useless as a and inertness or lability of the formed DOTA-cation model for DOTATATE. Finally, the capability for a metal complexes may represent a more predictive model. To our contaminant to compete with a radionuclide ( 90Y or 177Lu) knowledge, these data have not been published except for in the complexation with the macrocycle may not be cobalt and lutetium [3,19]. explained simply on the basis of the thermodynamic stability Finally, it is worth noting that the competition described in of the complexes. However, a significant correlation this study is based on a peptide/radionuclide molar ratio equal between the labeling yields and the ionic radii of the metal to 1. This ratio can sensibly differ from the ratio used in a contaminants was found, suggesting that factors related to typical radiopharmaceutical preparation for clinical applica- ion dimensions might play a role in their efficiency as tion. In fact, when the radiolabeling is performed in a nuclear competitors. Therefore, we can speculate on factors that may medicine facility, the molar ratio between DOTATATE and affect the labeling yields that are not accounted for by yttrium-90 or lutetium-177 is usually higher than 1, ranging thermodynamic stability constants: (a) different thermody- from 10 to 30 for 90Y and from 2 to 5 for 177 Lu, if a 177 Lu namic stabilities (formation constants) for metal-DOTA specific activity of 740 GBq/mg is considered, respectively. derivatives (and metal hydroxo species) at 25°C or 90°C, (b) Nonetheless, the results indicate that the chemical purity of kinetic effects in ligand metallation/demetallation processes commercial yttrium-90 and lutetium-177 chloride solutions is and (c) kinetic effects due to formation/precipitation/ an important factor for the selection of a suitable product dissolution of hydroxo species. since a formation yield N97% is theoretically achievable also To explain the temperature effect on thermodynamic for a radionuclide:ligand molar ratio of 1, in spite of the stabilities, we have to consider the relation between the reaction kinetics, if metal contaminants, such as Cu 2+, Zn 2+, logarithm of the complex formation constant and enthalpy at Co 2+, Ni 2+, Pb 2+ and Fe 3+, are sufficiently low. different temperatures: the more exothermic the M+ DOTA=M(DOTA) reaction, the less favored is the process 4. Conclusion at high temperatures, as found recently for lanthanide complexes with DTPA. In the absence of stability Many cations can compete with yttrium or lutetium in the constant values determined at temperatures higher than coordination reaction with DOTATATE. In this perspective 25°C, it is very difficult to correctly evaluate the speciation at study, a complete profile of Y- and Lu-DOTATATE 90°C. In fact, when slow kinetic processes are involved, it coordination yields in relation to the concentration of various might be expected that the equilibrium conditions at 90°C cations was obtained. These results allowed establishing, for are frozen upon cooling down the reacted sample. Therefore, a radionuclide:DOTATATE molar ratio equal to 1, the metal the speciation at the end of the heating cycle better reflects contaminants affecting the labeling yields. Moreover, we that at high temperature, instead of that at 25°C. established that the evaluation of the amount of species at The kinetics in metal-macrocycle formation can vary equilibrium using formation constants at 25°C does not highly among metals, being extremely fast in some cases, reflect the experimental quantities. However, due to the e.g., labile Cu(II), and very slow for some demetallation higher molar ratio of DOTATATE to radionuclides for process, e.g., Cu(II)-cyclam complex is very stable at very clinical radiopharmaceutical preparations, a lower interfer- low pH for days, even if complete removal of Cu 2+ from ence of these metals is expected. the macrocycle is expected from the determined stability At a fixed peptide amount, this study is crucial for constants. Therefore, contaminants with less stable estimating the admissible limits of individual metal impu- complexes of Y and Lu, if present in excess, can quickly rities that can be accepted in a solution of radioactive coordinate to the ligand and be removed. Finally, solid precursors for performing a successful radiolabeling. The phase kinetic rates (e.g., precipitates) are largely unpredict- results obtained herein are transferable, with a good able. This factor, along with the thermodynamics of approximation, to any other DOTA-conjugated biomolecule formation of insoluble or colloidal metal hydroxides, may used for binding the metals considered. affect the quantities of competing metal ions available for DOTA complexation. Acknowledgment A direct comparison of the numerical values is difficult due to the different molar ratios of Y 3+/Lu 3+ and ligands The present study was performed with the support of used in the studies presented in the literature [7-9]; our PerkinElmer. The authors want to thank L. Murray, B. Lynch results confirm that in the presence of cationic competitors, and A. Ketring for helpful discussions. the radionuclide-DOTATATE formation is highly dependent on the concentration of the individual cation. However, our References experimental data and computations are based on the assumption of a single competitor effect and are not able Chinol M, Paganelli G. Radionuclide peptide cancer therapy. New to predict the behavior of DOTATATE chelator when more York, United States: Taylor & Francis; 2007.  M. Asti et al. / Nuclear Medicine and Biology xx (2011) xxx–xxx 9 Stimmel JB, Stockstill ME, Kull FC. Yttrium-90 chelation properties Grassi E, Sghedoni R, Asti M, Fioroni F, Salvo D, Borasi G. Radiation of tetraazatetraacetic acid macrocycles, diethylenetriaminepentaacetic protection in 90Y-labelled DOTA-D-Phe 1-Tyr 3-octreotide prepara- acid analogues, and a novel terpyridine acyclic chelator. Bioconjugate tions. Nucl Med Commun 2009;30:176–82. Chem 1995;6:219–25. Urbano N, Papi S, Ginanneschi M, et al. Evaluation of a new biotin- Stimmel JB, Kull FC. Samarium-153 and lutetium-177 chelation DOTA conjugate for pretargeted antibody-guided radioimmunother- properties of selected macrocyclic and acyclic ligands. Nucl Med Biol apy (PAGRIT®). Eur J Nucl Med Mol 2007;34:68–77. 1998;25:117–25. Kumar K, Chang CA, Tweedle MF. Equilibrium and kinetic study of Anderegg G, Arnaud-Neu F, Delgado R, Felcman J, Popov K. Critical lanthanide complexes of macrocyclic polyamino carboxylates. Inorg evaluation of stability constants of metal complexes of complexones Chem 1993;32:587–93. for biomedical and environmental applications. IUPAC Technical Brücher E, Sherry AD. Stability and toxicity of contrast agents. In: Report. Pure Appl Chem 2005;77:1445–95. Merbach AE, & Toth E, editors. The chemistry of contrast agents in Chaves S, Delgado R, Silva JJRF. The stability of the metal complexes medical magnetic resonance imaging. Chirchester, UK: John Wiley & of cyclic tetra-aza tetra-acetic acids. Talanta 1992;39:249–54. Sons; 2001, pp. 243–79. Liu S, Edwards DS. Bifunctional chelator for therapeutic lanthanide Viola-Villegas N, Doyle RP. The coordination chemistry of 1,4,7,10- radiopharmaceuticals. Bioconjugate Chem 2001;12:7–34. tetraazacyclododecane-N,N',N'’,N'’'-tetraacetic acid (H4DOTA): Breeman WAP, de Jong M, Visser TJ, Erion JL, Krenning EP. structural overview and analyses on structure–stability relationships. Optimizing condition for radiolabelling of DOTA-peptides with 90Y, Coord Chem Rev 2009;253:1906–25. 111 In, and 177Lu at high specific activities. Eur J Nucl Med Mol Greenwood NN, Earnshaw A. Chemistry of the elements; 1991. Piccin Imaging 2003;30:917–20. Nuova Libraria-S.p.A. Padova, Italy. Eigner Henke K, Beran M, Srank J, Melichar F. Quality control of no- Heppeler A, Andrè JP, Buschmann I, et al. Metal-ion-dependent carrier added 90YCl3 by estimating the labelling efficiency using its biological properties of a chelator-derived somatostatin analogue for reaction with DOTATOC. J Radioanal Nuc Chem 2008;278:25–9. tumour targeting. Chem Eur J 2008;14:3026–34. Pawlak D, Korsak A, Mikolajczak R, Janota B, Karczmarczyk U, Heppeler A, Froidevaux S, Maecke HR, et al. Radiometal-labelled Jakubowska E. Comparative evaluation of therapeutic radiopharma- macrocyclic chelator-derivatised somatostatin analogue with superb ceutical. I.A.E.A. Technical Reports Series no. 458;13:217–32. tumor-targeting properties and potential for receptor-mediated internal Smith RM, Martell AE, Motekaitis RJ. NIST critically selected radiotherapy. Chem Eur J 1999;5:1974–81. stability constants of metal complexes database, V. 8.0, NIST, Tian G, Martin LR, Zhang Z, Rao L. Thermodynamic, spectroscopic, Gaithersburg, MD 20899. and computational studies of lanthanide complexation with diethyle- Pettit LD, Powell HKJ. The IUPAC stability constants database. netriaminepentaacetic acid: temperature effect and coordination London, UK: Royal Society of Chemistry; 2000. modes. Inorg Chem 2011;50:3087–96. Alderighi L, Gans P, Ienco A, Peters D, Sabatini A, Vacca A. Boiocchi M, Bonizzoni M, Fabbrizzi L, et al. Does a reinforced kinetic Hyperquad simulation and speciation (HySS): a utility program for the macrocyclic effect exist? The demetallation in strong acid of copper(II) investigation of equilibria involving soluble and partially soluble complexes with open and cyclic tetramines containing a piperazine species. Coord Chem Rev 1999;184:311–8. fragment. Chem Eur J 2004;10:3209–16. View publication stats