Feasibility of the radioastatination of a monoclonal antibody with astatine-211 purified by wet extraction

Article abstract of DOI:10.1002/jlcr.1543

Astatine-211, a most promising α-particle emitter for targeted radiotherapy, is generally obtained by high-temperature distillation. However, a liquid-liquid extraction procedure (wet extraction) has also been described. The purpose of this study was to develop and optimize the labelling of the stannylated-activated ester N-hydroxysuccinimidyl-meta-rimethylstannylbenzoate ester (MeSTB) with astatine-211 extracted in di-isopropylether (DIPE) in the presence of the oxidant N-chlorosuccinimide (NCS). The effect of final volume, incubation duration and NCS amounts on radiolabelling yield was studied. The best yields (85-90%) of N-hydroxysuccinimidyl-meta-[²¹¹At] astatobenzoate ester (SAB) were obtained with 20 nmol of MeSTB, 100 nmol of NCS in 120 μL of DIPE after 15 min. The astatine-211-labelled-activated ester was then used to radiolabel a monoclonal antibody (mAb). The labelling yield was 20-25% and the radiochemical purity was 97-99%. These results show that mAbs may be efficiently labelled with astatine-211 obtained by wet extraction, a fully automatable technique that may prove to be a useful alternative to dry distillation for high activity labelling of radiopharmaceuticals. Copyright

Full text of DOI:10.1002/jlcr.1543

Research Article
Received 12 June 2008,
Revised 18 August 20, 08
Accepted 19 August 2008
Published online 24 September 2008 in Wiley Interscience
(www.interscience.wiley.com) DOI: 10.1002/jlcr.1543
Feasibility of the radioastatination of a
monoclonal antibody with astatine-211
purified by wet extraction
a,b
Ã
Franc-ois Guerard,a Cyrille Alliot,a Marie Mougin-
¨
Mickael Bourgeois,
Degraef, Holisoa Rajerison, Patricia Remaud-Le Saec, Jean-Franc-ois
a
a
a
´
¨
a
a
a
a
´
Gestin, Franc-ois Davodeau, Michel Cherel, Jacques Barbet, and Alain
Faivre-Chauveta,b
Astatine-211, a most promising a-particle emitter for targeted radiotherapy, is generally obtained by high-temperature
distillation. However, a liquid–liquid extraction procedure (wet extraction) has also been described. The purpose of this
study was to develop and optimize the labelling of the stannylated-activated ester N-hydroxysuccinimidyl-meta-
trimethylstannylbenzoate ester (MeSTB) with astatine-211 extracted in di-isopropylether (DIPE) in the presence of the
oxidant N-chlorosuccinimide (NCS). The effect of final volume, incubation duration and NCS amounts on radiolabelling yield
was studied. The best yields (85–90%) of N-hydroxysuccinimidyl-meta-[211At]astatobenzoate ester (SAB) were obtained
with 20 nmol of MeSTB, 100 nmol of NCS in 120 mL of DIPE after 15 min. The astatine-211-labelled-activated ester was then
used to radiolabel a monoclonal antibody (mAb). The labelling yield was 20–25% and the radiochemical purity was 97–99%.
These results show that mAbs may be efficiently labelled with astatine-211 obtained by wet extraction, a fully automatable
technique that may prove to be a useful alternative to dry distillation for high activity labelling of radiopharmaceuticals.
Keywords: astatine-211; SAB; radioimmunotherapy; monoclonal antibody; a-particle
labelled with astatine-211 de-astatinate rapidly both in vitro and
in vivo.10,11
Introduction
Astatine-211 is generally purified by dry distillation.12 How-
Monoclonal antibodies (mAbs) directed against tumour-asso-
ciated antigens armed with various radionuclides are being
investigated as therapeutic agents for the treatment of
malignant diseases.1 mAbs radiolabelled with bÀ-emitting
radionuclides such as iodine-131 or yttrium-902 have demon-
strated their efficacy in the treatment of haematologic
malignant diseases. Radioimmunotherapy (RIT) with a-particle
emitters (a-RIT) is also being considered very actively.3 The
radiobiological properties of a-particles, due to a high ionization
potential per unit path length, are very attractive. The high
linear energy transfer of a-particles results in relatively short
effective path lengths in tissues and a-particles deliver a highly
localized radiation dose to malignant cells.4 This makes them
highly cytotoxic with a relative biologic effectiveness 5–20 times
that of bÀ-particles.5,6 Another advantage of a-particles,
ever, this technique has been described as affording variable
yields for the high activities required for clinical trials.13 Thus,
new approaches that could reproducibly provide high yields of
astatine-211 would be interesting. An alternative to dry
distillation has been described recently. In this technique,
instead of distilling astatine out of the irradiated target, the
layer of irradiated bismuth is dissolved in nitric acid and astatine
is extracted in di-isopropylether (DIPE).13 This technique is
referred to as wet extraction. The activated ester N-hydro-
xysuccinimidyl-meta-trimethylstannylbenzoate ester (MeSTB)
has been radiolabelled with astatine-211 obtained by wet
extraction, but so far mAb radiolabelling with this method has
not been reported because of hydrolysis of the N-hydoxysucci-
nimide-activated ester.13 The aim of the present study was to
compared with bÀ-particles, is that they exhibit
a low
dependence on dose rate and oxygen enhancement effects.7
Astatine-211 is the heaviest halogen and it is one of the most
promising a-emitters for medical applications. The half-life of
astatine-211 is relatively long compared with that of other
radionuclides available for a-RIT (T1/2 = 7.2 h). Its potential utility
for radiotherapy has been recognized 50 years ago.8,9 The main
difficulty of protein radiolabelling with astatine-211 is the
imperfect knowledge of the chemistry of astatine because there
is no stable isotope of this element. Moreover, proteins directly
a
´
Centre de Recherche en Cancerologie de Nantes-Angers, Inserm, Institut de
Biologie, Universite de Nantes, U892, 9 quai Moncousu, 44093 Nantes Cedex 1,
´
France
b
ˆ
Department of Nuclear Medicine, CHU de Nantes-Hotel Dieu, Nantes, France
¨
´
*Correspondence to: Mickael Bourgeois, Centre de Recherche en Cancerologie de
Nantes-Angers, Inserm, Institut de Biologie, Universite de Nantes, U892, 9 quai
Moncousu, 44093 Nantes Cedex 1, France.
E-mail: Mickael.Bourgeois@univ-nantes.fr, mickael.bourgeois@nantes.inserm.fr
´
J. Label Compd. Radiopharm 2008, 51 379–383
Copyright r 2008 John Wiley & Sons, Ltd.

M. Bourgeois et al.
investigate the feasibility of radiolabelling of an mAb with (at end of bombardment). The wet extraction procedure takes
astatine-211 obtained by wet extraction. Thus, the reaction about 1.5 h because concentrated nitric acid must be evapo-
conditions were re-investigated and wet-extracted astatine-211 rated to dryness and 70–90 % (decay corrected) of the activity
was coupled to the immunological vector using MeSTB and the is collected in DIPE. At the end of the procedure, 60–77 MBq of
radiochemical yield was optimized.
astatine is recovered in 500 mL of DIPE. Dry distillation
gives similar yields in less time. Halogenated solvents (such
as chloroform) are generally used to recover the astatine
from a PEEK capillary loop.12,14,15 However, methanol, which is
also used to collect astatine after dry distillation, is a better
solvent for SAB preparation16 as it minimizes radiolysis. After
wet extraction, astatine is obtained as a solution in DIPE
containing nitric acid, in which radiolysis did not appear to be a
problem.
Results and discussion
Irradiation of a bismuth-209 target by the cyclotron of the CNRS
´
in Orleans (CEMHTI, Orleans, France) as described in the
Experimental section routinely affords 100 MBq of astatine-211
´
[
125I]-iodine-labelled iodobenzoic acid and N-hydroxysuccini-
midyl-meta-iodobenzoate ester were synthesized for reference
(Figure 1). High-performance liquid chromatography (HPLC)
retention times were 9.1 and 14.7 min (Figure 2) and thin-layer
chromatography (TLC) Rf were 0.56 and 0.68, respectively. The
amounts of both iodinated products were too small to be
detected by UV light. A commercial [127I]-iodobenzoic acid
reference gave an HPLC retention time of 8.9 min and a TLC Rf of
0.56 detected by UV light.
[
211At]-astatobenzoic acid and N-hydroxysuccinimidyl-meta-
211At]astatobenzoate ester gave HPLC retention times of 9.2
[
and 14.9 min, respectively (Figure 3). The TLC Rf were the same
as that of the iodinated references. The amounts of both
astatine-labelled products were too small to be detected by UV
light.
Figure 1. Halogeno-destannylation reaction of MeSTB (top) and meta-(trimethyl-
stannyl)benzoic acid (bottom).
Figure 2. HPLC radiochromatogram of meta-iodobenzoic acid (left) and N-hydroxysuccinimidyl-meta-iodobenzoate ester (right). Reverse phase HPLC was performed
using a mbondapak C18 analytical column under isocratic conditions (0.1% trifluoroacetic acid in water/acetonitrile, 85:15, v/v). Measured retention times were 9.1 and
14.7 min, respectively. In the radiochromatogram of N-hydroxysuccinimidyl-meta-iodobenzoate ester, a small amount of meta-iodobenzoic acid is observed due to the
hydrolysis of activated ester.
Figure 3. HPLC radiochromatogram of meta-astatobenzoic acid (left) and N-hydroxysuccinimidyl-meta-astatobenzoate ester (right). Reverse phase HPLC was performed
under the exact same conditions as that used for the iodinated derivatives. Measured retention times were 9.2 and 14.9 min, respectively. In the radiochromatogram of N-
hydroxysuccinimidyl-meta-astatobenzoate ester, a small amount of meta-astatobenzoic acid is observed due to the hydrolysis of activated ester.
www.jlcr.org
Copyright r 2008 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2008, 51 379–383

M. Bourgeois et al.
Figure 4. Kinetics of astato-destannylation of MeSTB. The reaction mixture
contained 20 nmol of MeSTB and variable amounts of N-chlorosuccinimide (0, 2,
20, 100, 200 and 1000 nmol) in a 121 mL of final volume. Progress of the reaction
was analysed at selected time intervals by TLC and the percentage of activity
counted at the Rf of SAB was plotted as a function of time in the figure.
Figure 6. ITLC of astatine-labelled antibody. ITLC was performed on silica gel using
10% trichloroacetic acid in water as the mobile phase. Lane 1 shows the ITLC of
crude astatine-labelled antibody and lane 2 the same preparation after size-
exclusion purification. In this chromatographic condition, an IgG was retained on
the baseline and free astatine was eluted to the solvent front.
Figure 5. Radio-TLC on silica gel of astatine and SAB. TLC was performed using
chloroform/ethyl acetate (8:2) as the mobile phase. The different lanes show: N-
hydroxysuccinimidyl-meta-iodobenzoate ester used as a reference (1), free astatine
obtained by DIPE extraction (2), free astatine obtained by DIPE extraction
incubated with NCS (3) and N-hydroxysuccinimidyl-meta-astatobenzoate ester (4).
Figure 7. Size-exclusion HPLC of astatine-labelled antibody. The figure shows the
radiochromatogram of astatine-labelled antibody purified by size-exclusion
chromatography, with the major peak at
corresponding to that of the IgG.
a retention time of 17.0 min
We have analysed the effect of the quantity of oxidizing agent
(N-chlorosuccinimide, NCS) with a constant amount of MeSTB
(20 nmol). Using 100 mL of wet-extracted astatine-211, the best
SAB yield was obtained with a minimum of 100 nmol of NCS and
a minimum time of 5–10 min (Figure 4). It is interesting to note
that in the absence of NCS or using only a small amount
(2 nmol) of NCS, an astatinated species with an Rf similar to that
of SAB is rapidly formed. However, the product is unstable and
free astatine release is observed. Increasing the NCS amount
beyond 100 nmol did not result in higher yields of SAB. The best
SAB yields were obtained when the volume was kept below
120 mL. Using these parameters high repeatability was achieved
with SAB yields in the 85–90% range. To keep the hydrolysis of
SAB into astatobenzoic acid to a minimum (Figure 3), it
appeared quite important to stop the evaporation of the
reaction mixture just before complete drying.
concentrations as in the test experiments of SAB synthesis.
Under these conditions, we obtained, in 15 min, a yield of
85–90% (Figure 5) of SAB. After the evaporation of DIPE,
purification of the astatine derivative using a SepPak C18
cartridge followed by evaporation of the acetonitrile phase, in
30 min, 70–75% of the activity was recovered.
The conjugation of SAB to mAb was performed in 30 min with
a yield of 20–25% (Figure 5). After purification on NAP-5 size-
exclusion columns, the astatine-labelled IgG was collected in a
volume of 0.8–1.4 mL. The radiochemical purity of this astatine-
labelled mAb was 97–99%. (Figure 6) and the specific activity
was 30–37 MBq/mg with a global yield of 20–25%. The fast
protein liquid chromatography retention time was 17.03 min in
agreement with IgG retention time on the Superdex 200 column
(Figure 7). The radioastatination overall yield of the antibody
was in accordance with the results obtained using dry
distillation.15,17
For the radiolabelling of a mAb, we have used a volume of
400 mL of astatine solution in DIPE. The amounts of NCS and
MeSTB were increased accordingly to maintain the same final
J. Label Compd. Radiopharm 2008, 51 379–383
Copyright r 2008 John Wiley & Sons, Ltd.
www.jlcr.org

M. Bourgeois et al.
Briefly, 5 mL of saturated aqueous sodium chloride solution,
20 nmol of stannylated precursor in methanol/acetic acid 95:5,
100 nmol of NCS in methanol/acetic acid 95:5 and 2 mL (7.4 MBq)
of sodium [125I]iodide in 1 Â 10À5 M NaOH were incubated for
15 min at room temperature. A 10 mL aliquot of the reaction
mixture was diluted with 50 mL of methanol and 40 mL from this
solution was then analysed using analytical HPLC.
Experimental
Materials
NCS was purchased from Acros Organics (Geel, Belgium). All
other chemicals and solvents were obtained from Carlo Erba (Val
de Reuil, France) and were used without further purification.
Sodium [125I]iodide was obtained from Perkin Elmer (Boston,
MA, USA). MeSTB for radioiodination and radioastatination was
prepared as described.18 The mAb (MJ7/18) is a rat IgG2b mAb
directed against the CD105 antigen (endoglin).19 This mAb was
produced from an hybridoma purified by our colleagues at the
Yield
[
optimization
for
N-hydroxysuccinimidyl-meta-
211At]astatobenzoate ester production
The following experimental plan was used to analyse the
influence of kinetics, NCS concentration and volume of
radiolabelling reaction. A stock solution of NCS was prepared
by dissolving 66.7 mg of NCS in 50 mL of methanol/acetic acid
(95:5, v/v) to yield a final concentration of 100 nmol/mL.
Solutions of NCS at 20, 10, 2, 0.2 or 0 nmol/mL were prepared
by dilution in methanol/acetic acid (95:5, v/v). To 100 mL of 211At
in DIPE were added 10.9 mL of MeSTB (20 nmol) and 10 mL of the
different solutions of NCS. The yield of the reaction was
determined by TLC at 5, 10, 15, 30 and 60 min.
Three concentrations of 211At in DIPE were prepared:
postextraction 211At in DIPE (50 mL), 211At (50 mL) diluted with
same volume of DIPE, postextraction 211At (100 mL) in DIPE
diluted with three volumes of DIPE. Astatine-211 in DIPE was
incubated with 10.9 mL of MeSTB (20 nmol) and 10 mL of NCS
(100 nmol). The final volumes were, respectively, 71, 121 and
421 mL. The yield of the reaction was determined by TLC at
15 min.
´
Centre de Recherche en Cancerologie de Nantes-Angers
(Nantes, France).
HPLC was performed using a Waters HPLC system (Saint
Quentin en Yvelines, France), equipped with a Waters 486
Tunable Absorbance Detector and a Packard Bioscience Flow
Scintillation Analyser 150 TR (Meriden, CT, US). Data analysis was
carried out using Waters Empower data acquisition and analysis
software. For the iodine and astatine organic benzoate
derivatives, we used a Waters mbondapak C18 (3.9 Â 300 mm;
5 mm) analytical column. The isocratic mobile phase was
composed of trifluoroacetic acid (0.1%) in water/acetonitrile
(85:15, v/v). The mobile phase was filtered though a 0.45 mm
Whatman (Maidstone, UK) nylon membrane filter. Outgassing
was performed using a Waters inline degasser AF. Flow rate was
set to 1 mL/min. The analysis time was 30 min. UV absorbance
was monitored at 254 nm. For the control of the astatine-
labelled antibody, we used a Superdex 200 10/300GL column
(GE Healthcare Bio-science, Uppsala, Sweden). The isocratic
mobile phase was 0.1 M phosphate buffer, pH 7.2, at a flow rate
of 1 mL/min with an analysis time of 35 min.
Radiolabelling of an mAb with N-hydroxysuccinimidyl-meta-
[
211At]astatobenzoate ester
Radio-TLC was carried out on precoated silica gel 60 F254 TLC
plastic sheets (Merck, Darmstadt, Germany) with chloroform/
ethyl acetate 8:2 as mobile phase. Instant TLC on silica gel (ITLC)
was carried out on precoated silica gel on glass fibre sheets (Pall
Life Science, New York, USA) with trichloroacetic acid in water
(10%) as mobile phase. Radio-TLC plates were examined using a
Typhoon 9410 Variable Mode Imager (GE Healthcare Bio-
science).
In the first step, 400 mL of astatine-211 in DIPE was reacted with
400 nmol of NCS and 80 nmol of MeSTB in 63.6 mL of methanol/
acetic acid 95:5 during 15 min at room temperature. The yield of
the reaction was determined by TLC. The reaction mixture was
evaporated to 50 mL under a stream of nitrogen at room
temperature, loaded on a SepPak C18 column (Waters) and
eluted with 2 mL of trifluoroacetic/water (0.1%) and second with
2 mL of acetonitrile. The acetonitrile phase was dry evaporated
under a stream of nitrogen at 401C.
Cyclotron irradiation
The coupling reaction was performed by the addition of
200 mL of mAb (IgG MJ7/18, 0.4 mg/mL)) in carbonate buffer
(300 mM, pH 8.6) to the dry residue of SAB. The yield of the
radiolabelling was determined by ITLC.
Finally, the astatine-labelled antibody was purified using a
NAP-5 column (Amersham Biosciences, Uppsala, Sweden) with a
phosphate-buffered saline as eluent. The radiochemical purity of
the antibody was determined by ITLC and Superdex G200 HPLC.
Astatine-211 was produced at the CNRS cyclotron of CEMHTI
using the 209Bi(a,2n)211At reaction by bombarding a 240 mm
thick natural 209Bi layer on copper target with a 1.95–2.15 mA
particle beam of 28 MeV a-particles. The irradiation lasted for 2 h.
Separation of 211At from the bismuth target
The irradiated target was placed in a beaker and 2 mL of 65%
nitric acid was added drop-wise on the irradiated target to
completely dissolve the bismuth layer. After 10–15 min, the acid
solution was evaporated to dryness and, after cooling, the
residue was re-dissolved in 2 mL of 32% nitric acid. Astatine-211
was then extracted in DIPE (2 Â 500 mL). The extracts were
combined and the total harvested activity was measured in an
ACAD 2000 ionization chamber (Lemer Pax, Carquefou, France).
Conclusion
Wet harvesting of astatine has been discussed previously, but
antibody radiolabelling with SAB obtained using this procedure
has not yet been reported.13 In this paper, we have re-evaluated
the methodology and reproducibly obtained astatine-labelled
mAb in a 20–25% yield based on the astatine activity involved
with a typical specific activity of 26–44 MBq/mg. These results
could be improved by using higher astatine activities and higher
antibody concentrations. In this paper, we have studied the
Iodine labelling Procedure
Radioiodination of MeSTB and meta-trimethylstannylbenzoic impact of different factors on the tin electrophilic substitution.
acid was performed using NCS as the oxidizing agent (Figure 1). The overall time required for the labelling and the SAB
www.jlcr.org
Copyright r 2008 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2008, 51 379–383

M. Bourgeois et al.
purification remains to be further optimized. An advantage of
the wet extraction procedure is that it requires no specific
equipment in contrast to dry distillation12 and that it could be
easily automated. Ultimately wet extraction could be a useful
alternative to dry distillation to obtain reproducible yields with
high activities of astatine.
¨
[5] T. Back, H. Andersson, C. R. Divgi, R. Hultborn, H. Jensen, S.
Lindegren, S. Palm, L. Jacobsson, J. Nucl. Med. 2005, 46,
2061–2067.
[6] K. Vandenbulcke, P. De Vos, F. Offner, J. Philippe, C. Apostolidis, R.
´
Molinet, T. K. Nikula, K. Bacher, V. de Gelder, A. Vral, C. Lahorte, H.
Thierens, R. A. Dierckx, G. Slegers, Eur. J. Nucl. Med. Mol. Imaging.
2003, 30, 1357–1364.
[7] T. E. Wheldon, J. A. O’Donoghue, Int. J. Radiat. Biol. 1990, 58, 1–21.
[8] W. L. Hughes, D. Gitlin, Am. Soc. Biol. Chem. 1955, 14, 229.
[9] P. W. Durbin, C. W. Asling, M. E. John, Radiat. Res. 1958, 9,
378–397.
Acknowledgement
The authors thank their colleagues at the CEMHTI cyclotron in [10] C. Aaij, W. R. J. M. Tschroots, L. Lindner, T. E. W. Feltkamp, Int. J.
Appl. Radiat. Isot. 1975, 26, 25–30.
´
Orleans, France, for astatine production and Alexander Yorda-
nov for the helpful discussions. This work was supported by the
[11] A. T. M. Vaughan, J. H. Fremlin, Int. J. Nucl. Med. Biol. 1978, 5,
229–230.
[12] S. Lindegren, T. Back, H. Jensen, Appl. Radiat. Isot. 2001, 55,
157–160.
[13] A. T. Yordanov, O. Pozzi, S. Carlin, G. Akabani, B. Wieland, M. R.
Zalutsky, J. Radioanal. Nucl. Chem. 2004, 262, 593–599.
[14] M. R. Zalutsky, X. G. Zhao, K. L. Alston, D. Bigner, J. Nucl. Med. 2001,
42, 1508–1515.
´
ˆ
Canceropole Grand Ouest and grants from the French Agence
Nationale de la Recherche (grant ANR-PCV VECRIT, A. F.-C.) and
from the EC (FP7 collaborative program TARCC, J. B.).
[15] A. Walte, S. S. R. Sriyapureddy, Z. Korkmaz, D. Krull, O. Bolte, M.
Hofmann, G. J. Meyer, W. H. Knapp, J. Pharm. Pharm. Sci. 2007, 10,
277s–285s.
References
[1] R. M. Sharkey, D. M. Goldenberg, CA Cancer J. Clin. 2006, 56, [16] O. R. Pozzi, M. R. Zalutsky, J. Nucl. Med. 2005, 46, 1393–1400.
226–243.
[2] S. I. Park, O. W. Press, Curr. Opin. Hematol. 2007, 14, 632–638.
[17] S. Palm, T. Back, I. Claesson, A. Danielson, J. Elgqvist, S. Frost, R.
Hultborn, H. Jensen, S. Lindegren, L. Jacobson, Int. J. Radiat. Oncol.
Biol. Phys. 2007, 69, 572–579.
´
´ ´
[3] M. Cherel, F. Davodeau, F. Kraeber-Bodere, J. F. Chatal, Q. J. Nucl.
Med. Mol. Imaging 2006, 50, 322–329.
[4] M. R. Zalutsky, O. R. Pozzi, Q. J. Nucl. Med. Mol. Imaging 2004, 48,
289–296.
[18] J. Koziorowski, C. Henssen, R. Weinreich, Appl. Radiat. Isot. 1998,
49, 955–959.
[19] A. Z. Ge, E. C. Butcher, Gene 1994, 138, 201–206.
J. Label Compd. Radiopharm 2008, 51 379–383
Copyright r 2008 John Wiley & Sons, Ltd.
www.jlcr.org