Occasional survey
`
`Optimising conditions for radiolabelling of DOTA-peptides
`with 90Y, 111In and 177Lu at high specific activities
`
`Wouter A. P. Breeman1, Marion de Jong1, Theo J. Visser2, Jack L. Erion3, Eric P. Krenning1, 2
`
`1 Department of Nuclear Medicine, Erasmus MC Rotterdam, Rotterdam, The Netherlands
`2 Department of Internal Medicine, Erasmus MC Rotterdam, Rotterdam, The Netherlands
`3 BioSynthema, St. Louis, Mo., USA
`
`Published online: 4 April 2003
`© Springer-Verlag 2003
`
`Abstract. DOTA-conjugated peptides, such as [DOTA0,
`Tyr3]octreotide (DOTATOC) and [DOTA0,Tyr3]octreo-
`tate (DOTA-tate), can be labelled with radionuclides
`such as 90Y, 111In and 177Lu. These radiolabelled somato-
`statin analogues are used for peptide receptor radionu-
`clide therapy (PRRT). Radioligands for PRRT require
`high specific activities. However, although these radio-
`nuclides are produced without addition of carrier, con-
`taminants are introduced during production and as decay
`products. In this study, parameters influencing the kinet-
`ics of labelling of DOTA-peptides were investigated and
`conditions were optimised to obtain the highest achiev-
`able specific activity. The effects of contaminants were
`systematically investigated, concentration dependently,
`in a test model mimicking conditions for labelling with
`minimal molar excess of DOTA-peptides over radionu-
`clide. Kinetics of labelling of DOTA-peptides were opti-
`mal at pH 4–4.5; pH <4 strongly slowed down the kinet-
`ics. Above pH 5, reaction kinetics varied owing to the
`formation of radionuclide hydroxides. Labelling with
`90Y and 177Lu was completed after 20 min at 80°C, while
`labelling with 111In was completed after 30 min at
`100°C. The effects of contaminants were systematically
`categorised, e.g. Cd2+ is the target and decay product of
`111In, and it was found to be a strong competitor with
`111In for incorporation in DOTA. In contrast, Zr4+ and
`Hf4+, decay products of 90Y and 177Lu, respectively, did
`not interfere with the incorporation of these radionucl-
`ides. The following conclusions are drawn: (a) DOTA-
`peptides can be radiolabelled at high specific activity;
`(b) reaction kinetics differ for each radionuclide; and (c)
`reactions can be hampered by contaminants, such as tar-
`get material and decay products.
`
`Keywords: DOTA – Radiolabelling – Specific activity
`
`Wouter A. P. Breeman (✉)
`Department of Nuclear Medicine, Erasmus MC Rotterdam,
`Dr. Molewaterplein 40, 3015 GD Rotterdam, The Netherlands
`e-mail: w.a.p.breeman@erasmusmc.nl
`Tel.: +31-10-4635317, Fax: +31-10-4635997
`
`Eur J Nucl Med Mol Imaging (2003) 30:917–920
`DOI 10.1007/s00259-003-1142-0
`
`Introduction
`
`DOTA-conjugated peptides, such as the stable somatosta-
`tin analogues [DOTA0,Tyr3]octreotate (DOTA-tate) and
`[DOTA0,Tyr3]octreotide (DOTATOC), can be readily la-
`belled with radionuclides such as 90Y, 111In and 177Lu. In
`order for these radiolabelled peptides to be successfully
`used in peptide receptor radionuclide therapy (PRRT) [1,
`2, 3, 4, 5, 6], high specific activities (SAs) are required. A
`number of biological factors dictate the need for a high
`SA. First, for in vivo use the amount of (radio)ligand that
`can be administered is limited by affinity and the amount
`of receptors. Above the optimal dose a further increase in
`ligand will increase the competition between unlabelled
`and labelled ligand for the same receptor and thus lower
`the uptake of radiolabel into receptor-positive tissue [4].
`Second, for peptides
`that display pharmacological
`(side)effects, such as DOTA-substance P or DOTA-bom-
`besin, only very small quantities of peptides may be toler-
`ated. For the latter peptide, the amount that can be admin-
`istered intravenously is limited to 0.1 nmol per minute
`[7]. Therefore, a high SA will reduce the total peptide
`amount to be administered. Third, endocytotic mecha-
`nisms that affect the cellular internalisation of peptides
`may become desensitised at high peptide concentrations
`[8], resulting in lower uptake of radiolabel into target tis-
`sue. Additionally, in vitro investigations aimed at measur-
`ing receptor-binding affinities require low concentrations
`of these radioligands (e.g. 10(cid:60)10 M) in order to measure
`receptor–ligand interactions accurately. Unfortunately,
`the need for high SA is often compromised by conflicting
`radiochemical parameters that determine reaction rates
`and yields, i.e. the rate of formation of the metal-DOTA
`complexes increases with pH [9], but the solubility of
`In3+, Y3+ and Lu3+ decreases with pH owing to formation
`
`European Journal of Nuclear Medicine and Molecular Imaging Vol. 30, No. 6, June 2003
`
`Evergeen Ex. 1020
`1 of 4
`
`

`

`918
`
`Table 1. Production methods, target materials, decay products, physical constants and maximal SA for the radionuclides considered
`
`Production method
`
`Target
`Decay product
`
`Physical constants
`t1/2 (days)
`pmoles per mCi
`
`Maximal SA (mCi per nmol)
`Theorya
`In practiceb
`
`57Co
`
`Reactor
`(n, pn)
`58Ni
`57Fe
`
`270.9
`2080
`
`0.5
`n.d.
`
`67Ga
`
`90Y
`
`111In
`
`Cyclotron
`(p, 2n)
`68Zn
`67Zn
`
`3.26
`25
`
`40
`10
`
`Generator
`
`90Sr
`90Zr
`
`2.67
`20.5
`
`49
`11
`
`Cyclotron
`(p, 2n)
`112Cd
`111Cd
`
`2.83
`21.5
`
`46
`22
`
`177Lu
`
`Reactor
`(n)
`176Lu
`177Hf
`
`6.71
`51.4
`
`19
`3
`
`Labelling conditions with 57Co and 67Ga were not optimised, but
`taken to be identical to those of 111In: 30 min at 100°C
`SA, Specific activity, expressed as mCi per nmol; n.d., not deter-
`mined
`a Since 1 nmol DOTA can incorporate 1 nmol (radio)nuclide, this
`number indicates the maximal theoretical SA of the radiolabelled
`DOTA-peptides
`
`of hydroxides [10]. Although recently labelling of DOTA
`analogues with 90Y and 177Lu was reported at pH 7–8 [11,
`12], we encountered the above-mentioned solubility
`problems when high concentrations of these radionucl-
`ides were used at pH 7–8. Therefore we decided to per-
`form this study at pH 5 or lower.
`The studies presented here were undertaken to deter-
`mine the optimal conditions for radiolabelling DOTA-
`peptides, using the radionuclides 90Y, 111In and 177Lu and
`DOTATOC and DOTA-tate as model reactants. We in-
`vestigated parameters that influence reaction kinetics
`and radiochemical yields in order to define conditions
`that result in maximal achievable SAs. This report also
`summarises the effects of ever-present contaminants,
`such as nuclides formed by decay of the radionuclides.
`
`Materials and methods
`
`Ligands and radionuclides. DOTA-peptides were dissolved in
`0.01–0.05 M acetic acid in Milli-Q water. As the buffer, 25 mM
`sodium ascorbate (quencher) in 50 mM sodium acetate was used.
`90Y was from Pacific Northwest National Laboratory (Richland,
`Wash., USA) and from MDS Nordion (Fleurus, Belgium). 177Lu
`was from Missouri University Reactor Research (MURR, St.
`Louis, Mo., USA) and from NRG (Petten, The Netherlands).
`57CoCl2 in 0.1 M HCl was from MDS Nordion (Vancouver, BC,
`Canada). 111InCl3 was from Mallinckrodt Medical (Petten, The
`Netherlands). All these radionuclides were delivered in 0.01–0.2 N
`HCl. 67GaCl3 in 0.7–0.8 M HCl was from Mallinckrodt Medical
`(Petten, The Netherlands) and from MDS Nordion (Vancouver,
`BC, Canada) in 0.05 M HCl.
`
`b Highest value achieved; this implies a ratio of DOTA over radio-
`nuclide [practice (b) over theory(a)] of 4, 4 1/2 and 2 1/4 for 67Ga,
`90Y and 111In, respectively
`
`using double-sealed plastic reaction tubes (PCR thermocycler
`tubes, max volume 125 μl, MoBiTec, ITK Diagnostics, Uithoorn,
`The Netherlands). Heating was performed in a temperature-con-
`trolled heating block (Grant, Fisher Scientific, Zoeterwoude, The
`Netherlands). The pH of the reaction mixture was measured after
`the reaction. All experiments were carried at least in duplicate, us-
`ing at least three different production batches of the radionuclides
`from the suppliers mentioned above. All chemicals were pur-
`chased from Aldrich Chemicals (Zwijndrecht, The Netherlands),
`and were of the highest analytical grade available. The radiochem-
`ical purity of the radiolabelled DOTATOC and DOTA-tate was
`studied at room temperature in the presence of 4 mM DTPA pH 5.
`High-performance liquid chromatography and instant thin-layer
`chromatography were performed as described previously (see [13]
`and [14] respectively)..
`
`Maximal achievable SA, and effects of contaminants. In order to
`determine the maximal achievable SA ((cid:42) 98% incorporation of
`67Ga, 90Y, 111In or 177Lu), radiolabelling was performed using a
`constant amount of the radionuclides and increasing amounts of li-
`gand. To investigate the effects of the contaminants present, ex-
`periments were performed adding known amounts of contami-
`nants to the reaction vial at the start of the radiolabelling. For all
`radionuclides considered, the target material and the decay prod-
`ucts are summarised in Table 1, and all were tested concentration
`dependently. The starting conditions in the absence of added con-
`taminants were chosen to be critical, using a low mol/mol ratio of
`DOTA over radionuclide.
`
`Results
`
`Reaction kinetics
`
`Reaction conditions: the effects of pH and temperature. The ex-
`periments were performed in small volumes (typically 40–75 μl)
`
`Reaction kinetics with 90Y, 111In and 177Lu were found to
`be optimal at pH 4–4.5, with a steep decrease at lower
`
`European Journal of Nuclear Medicine and Molecular Imaging Vol. 30, No. 6, June 2003
`Evergeen Ex. 1020
`2 of 4
`
`

`

`919
`
`Fig. 1. Formation of 177Lu-DOTATOC as a function of pH after
`20 min at 80°C, as measured by the % incorporation of the radio-
`nuclide. Similar results were found with 90Y and with DOTA-tate
`as ligand
`
`Fig. 2. Formation of radiolabelled DOTATOC at pH 4 as a func-
`tion of time of incubation, as measured by the % incorporation of
`the radionuclide. Similar results were found with DOTA-tate as
`ligand. +, 177Lu at 80°C; ▼, 177Lu at 100°C; ■, 90Y at 80°C; ■,
`90Y at 100°C; ●, 111In at 80°C; ❍, 111In at 100°C
`
`pH and a slow decrease at higher pH (Fig. 1). In addi-
`tion, % incorporation of 111In and 177Lu at pH (cid:42)5 became
`non-reproducible: after centrifugation of these reaction
`vials, precipitation was found. The reaction kinetics were
`also found to be time- and temperature-dependent: reac-
`tions were complete with 90Y and 177Lu after 20 min at
`80°C, and with 111In after 30 min at 100°C (Fig. 2). Ta-
`ble 1 shows the highest achieved SAs of 67Ga, 90Y and
`111In, implying a mol/mol ratio of DOTA over radionu-
`clide of 4, 4 1/2 and 2 1/4, respectively. For 177Lu the
`mol/mol ratio was even 1.2 (see also Discussion). We
`were still able to label at high SAs 2 weeks after the pro-
`duction of 177Lu (data not shown).
`
`Effects of contaminants of maximal achievable SA
`
`To validate the test model the effect of the addition of
`unlabelled Y, In and Lu to the corresponding radionu-
`clide was investigated. As expected, the dilution of the
`SA of the radionuclide decreased its incorporation in the
`DOTA-chelator in a concentration-dependent manner
`(Fig. 3, Table 2). The addition of nuclides such as Hf, Zr
`and Sr had no effect on the % incorporation of the radio-
`nuclides, indicating that these nuclides are not competi-
`tors under these reaction conditions. In contrast, the ef-
`
`Fig. 3. Effects of contaminants on the incorporation of 177Lu in
`DOTA-tate by the controlled addition of non-radioactive nuclides
`
`Table 2. Effects of different concentrations of metal ions, as con-
`taminants in the reaction vial, on incorporation of radionuclides in
`DPTA-ligand
`
`0
`
`Ag+
`Hf4+
`Hg2+
`Sr2+
`Zr4+
`
`+
`
`Ga3+
`Y3+
`
`++
`
`Cd2+
`Co2+
`Cu2+
`In3+
`Fe2+
`Lu3+
`Ni2+
`Zn2+
`
`0, (cid:41)10% at 10 μM; +, (cid:42)10% at 1–10 μM; ++, (cid:42)10% at 1 μM
`
`fect of addition of nuclides such as Fe and Cd clearly
`showed that they are strong competitors for the incorpo-
`ration of radionuclide in the DOTA-chelator, as shown in
`Fig. 3 and Table 2. Table 1 also contains data on label-
`ling experiments with 57Co and 67Ga using the same con-
`ditions as those for 111In, 30 min at 100°C. With the
`Mallinckrodt-produced 67Ga, pH was difficult to control
`since the radionuclide is delivered in 0.7–0.8 M HCl.
`More importantly, the [Zn] (Zn is the target and decay
`product of 67Ga, see Table 1) was measured by ICP (data
`not shown) and found to exceed [Ga] frequently by
`>100-fold, with a dramatic effect on achieving high SAs.
`
`Discussion
`
`We recently reported that Cd is a strong competitor for
`111In incorporation in the DOTA-chelator, and since
`111Cd is formed from decaying 111In, the highest SA is
`achieved immediately after the production of 111In [15].
`Analogously, achieving a high SA with 67Ga will be very
`difficult owing to the [67/68Zn], present from target (68Zn)
`
`European Journal of Nuclear Medicine and Molecular Imaging Vol. 30, No. 6, June 2003
`
`Evergeen Ex. 1020
`3 of 4
`
`

`

`920
`
`and formed during decay. Even if the [68Zn] is low or ze-
`ro at the end of production of 67Ga, after one half-life of
`67Ga the [67Ga]=[67Zn].
`This study provides insights into the effects of con-
`taminants in daily radiolabelling practice when high SAs
`are required. It also improves the interpretation of the
`concentrations of contaminants as mentioned in the data
`sheet provided by the manufacturer of the radionuclides.
`For instance, the data sheet of 90Y (MDS Nordion) states
`that the maximal concentration of Zn in 90Y will not ex-
`ceed the level of 30 μg per Ci of 90Y. If this is so, howev-
`er, it implies that the mol/mol ratio for Zn will be more
`than 20 times that for Y; achieving a high SA would then
`not be possible.
`In Table 1 the highest achieved SAs of 67Ga, 90Y, 111In
`and 177Lu are presented, and imply a mol/mol ratio of
`DOTA over radionuclide of 4, 4 1/2, 2 1/4 and 6, respec-
`tively. This reflects the reaction conditions, including
`pH, temperature and the use of very pure reactants. From
`Table 2 it can be seen that Hf is not a competitor for
`177Lu under our reaction conditions. This implies that the
`ingrowth of Hf has no consequences for the maximal
`achievable SA. The rate of incorporation is >99%, even
`at a mol/mol ratio (DOTA-peptide over 176+177Lu) of 1.2.
`In addition, a high SA can still be achieved 2 weeks after
`the production of 177Lu, confirming that Hf is not a com-
`petitor for Lu in the incorporation in DOTA. Although
`stability constants of DOTA with many nuclides are
`available in the NIST database [16], data on reaction ki-
`netics are scarce. However, from the data presented here
`it can be concluded that the reaction kinetics of Lu, Y
`and In with DOTA are in the order: Lu>Y>>In.
`In conclusion, DOTA-peptides can be radiolabelled at
`high SA. Reaction kinetics differ for each radionuclide:
`with 90Y and 177Lu, conditions were optimal at pH 4–4.5
`and the reactions were complete after 20 min at 80°C,
`while labelling with 111In was completed after 30 min at
`100°C. Reactions can be hampered by contaminants,
`such as target material and decay products.
`
`Acknowledgements. The technical assistance of Deborah Thoon-
`sen and Erik de Blois is greatly acknowledged.
`
`References
`
`1. Otte A, Jermann E, Behe M, et al. DOTATOC: a powerful new
`tool for receptor-mediated radionuclide therapy. Eur J Nucl
`Med 1997; 24:792–795.
`2. Erion J, Bugaj JE, Schmidt MA, Wilhelm RR, Srinivasan A.
`High radiotherapeutic efficacy of [Lu-177]-DOTA-Y3-octreo-
`tate in a rat tumor model. J Nucl Med 1999; 40:223p.
`3. Krenning EP, Valkema R, Kooij PP, et al. The role of radioac-
`tive somatostatin and its analogues in the control of tumor
`growth. Recent Results Cancer Res 2000; 153:1–13.
`4. Breeman WA, de Jong M, Kwekkeboom DJ, et al. Somatosta-
`tin receptor-mediated imaging and therapy: basic science, cur-
`rent knowledge, limitations and future perspectives. Eur J
`Nucl Med 2001; 28:1421–1429.
`5. Kwekkeboom D, Krenning EP, de Jong M. Peptide receptor
`imaging and therapy. J Nucl Med 2000; 41:1704–1713.
`6. Kwekkeboom DJ, Bakker WH, Kooij PP, et al. [177Lu-
`DOTA0Tyr3]octreotate: comparison with [111In-DTPA0]octreo-
`tide in patients. Eur J Nucl Med 2001; 28:1319–1325.
`7. Jansen JB, Jebbink MC, Douglas BR, Lamers CB. Effect of
`loxiglumide (CR-1505) on bombesin- and meal-stimulated
`plasma cholecystokinin in man. Eur J Clin Pharmacol 1990;
`38:367–370.
`8. Bunemann M, Hosey MM. G-protein coupled receptor kinases
`as modulators of G-protein signalling. J Physiol 1999; 517:5–23.
`9. Szilagyi E TE, Kovacs Z, Platzek J, Radüchel B, Brücher E.
`Equilibria and formation kinetics of some cyclen derivative
`complexes of lanthanides. Inorganica Chemica Acta 2000;
`298:226–234.
`10. Moerlein SM Welch MJ. The chemistry of gallium and indium
`as related to radiopharmaceutical production. Int J Nucl Med
`Biol 1981; 8:2772–2787.
`11. Kukis DL, DeNardo SJ, DeNardo GL, O'Donnell RT, Meares
`CF. Optimized conditions for chelation of yttrium-90-DOTA
`immunoconjugates. J Nucl Med 1998; 39:2105–2110.
`12. Liu S, Cheung E, Ziegler MC, Rajopadhye M, Edwards DS.
`(90)Y and (177)Lu labeling of a DOTA-conjugated vitronectin
`receptor antagonist useful for tumor therapy. Bioconjug Chem
`2001; 12:559–568.
`13. Bakker WH, Krenning EP, Breeman WA, et al. In vivo use of a
`radioiodinated somatostatin analogue: dynamics, metabolism,
`and binding to somatostatin receptor-positive tumors in man J
`Nucl Med 1991; 32:1184–1189 [published erratum appears in
`J Nucl Med 1991; 32:1999] [see comments].
`14. Smith-Jones PM, Stolz B, Albert R, et al. Synthesis and char-
`acterisation of [90Y]-Bz-DTPA-oct: a yttrium-90- labelled oct-
`reotide analogue for radiotherapy of somatostatin receptor-
`positive tumours. Nucl Med Biol 1998; 25:181–188.
`15. Breeman W, Erion JL, Bakker, WH, Visser, TJ, Krenning, EP,
`de Jong, M. Optimizing conditions for radiolabeling of
`DOTA-peptides with90Y, 111In, and 177Lu at high specific ac-
`tivities. Eur J Nucl Med 2002; 29:P690.
`16. NIST. NIST critically selected stability constants of metal
`complexes. A.E. Martell, ed.
`
`European Journal of Nuclear Medicine and Molecular Imaging Vol. 30, No. 6, June 2003
`Evergeen Ex. 1020
`4 of 4
`
`