Synthesis of (E)-2,3′,4,5′-tetramethoxy[2-11C] stilbene

Article abstract of DOI:10.1002/jlcr.1439

In this paper, we describe the radiosynthesis of the compound (E)-2,3′,4,5′-tetramethoxy[2-¹¹C]stilbene, a potential, universal tumour positron emission tomography imaging agent. The production of (E)-2,3′,4,5′-tetramethoxy[2-¹¹C]sti

Full text of DOI:10.1002/jlcr.1439

Journal of Labelled Compounds and Radiopharmaceuticals
J Label Compd Radiopharm 2007; 50: 1206–1210.
Published online 19 September 2007 in Wiley InterScience
(www.interscience.wiley.com). DOI: 10.1002/jlcr.1439
JLCR
Research Article
Synthesis of (E)-2,30,4,50-tetramethoxy[2-11C]stilbene
LUTZ SCHWEIGER*, STUART CRAIB, ANDREW WELCH and TIM A. D. SMITH
John Mallard Scottish PET Centre, School of Medical Sciences, University of Aberdeen, Foresterhill, Aberdeen AB25 2ZD, UK
Received 30 March 2007; Revised 26 June 2007; Accepted 5 July 2007
Abstract: In this paper, we describe the radiosynthesis of the compound (E)-2,30,4,50-tetramethoxy[2-11C]stilbene, a
potential, universal tumour positron emission tomography imaging agent. The production of (E)-2,30,4,50-
11
tetramethoxy[2-11C]stilbene was carried out via C-methylation of (E)-2-(hydroxy)-30,4,50-trimethoxystilbene by
using [11C]methyl trifluoromethanesulfonate ([11C]methyl triflate). (E)-2,30,4,50-tetramethoxy[2-11C]stilbene was
obtained with a radiochemical purity greater than 95% in a 20 ꢀ 2% decay-corrected radiochemical yield, based
upon [11C]carbon dioxide. Synthesis, purification and formulation were completed on an average of 30 min following
the end of bombardment (EOB). The specific radioactivity obtained was 1.9 ꢀ 0.6 GBq/mmol at EOB. Copyright #
2007 John Wiley & Sons, Ltd.
Keywords: 2,30,4,50-tetramethoxy[2-11C]stilbene; CYP1B1; PET
Introduction
the cytochrome P450 (CYP) family is the protein P450
1B1, which is a major oestrogen (E2) 4-hydroxylase
and is involved in the metabolic activation of several
polycyclic aromatic hydrocarbon carcinogens including
benzo(a)pyrene, dibenzo(a,l)pyrene, 7,12-dimethylben-
z(a)anthracene and 5-methyl chrysene.7 Although
CYP1B1 mRNA has been found in a number of normal
human tissues, the presence of the protein appears to
be specific to tumour tissue.8 Using immunohisto-
chemistry, Murray et al.9 found that CYP1B1 was
absent in all 130 samples of normal human tissue
from 16 different anatomical regions. In contrast,
CYP1B1 protein was present in 122 of 127 samples of
malignancies from each of the 16 anatomical regions.
Further, immunoblotting and immunochemistry failed
to reveal the presence of native CYP1B1 protein in
human liver8 and only very low levels in cells prepared
from non-neoplastic breast tissue,10,11 but immuno-
blotting12 and immunochemistry9 consistently demon-
strated the presence of CYP1B1 in human breast
tumours. In 2002, a number of inhibitors of CYP1B1
have been developed,13 the most selective of which is
(E)-2,30,4,50-tetramethoxystilbene (TMS). Kim et al. dis-
covered that TMS inhibited P450 1B1 with a 50-fold
selectivity over P450 1A1 and 500-fold selectivity over
P450 1A2-, the other two members of the CYP enzyme
family.13 This inhibitor is metabolized by P450 1B1 at a
very slow rate, so that the molecule can be labelled in
Over the last decade positron emission tomography
(PET), using the glucose metabolism-imaging agent
2-deoxy-2-[18F]fluoro-d-glucose (FDG), started to play
a major role in the management of patients, especially
in oncology.1 Although FDG-PET is invaluable in the
imaging of a number of tumour types, it is not suitable
for detection of all tumours due to the high uptake of
FDG by normal tissues,2,3 inability to differentiate
some low/medium grade tumours from benign le-
sions,4 limited use in the diagnosis of some recurrent
tumours,5 as well as accumulation of FDG in inflam-
mation making it indistinguishable from tumour tis-
sue. Hence, other PET tracers with greater tumour
specificity need to be developed. Malignant tumour
cells can abnormally express or overexpress proteins
including enzymes which are consequently potential
targets for imaging with radiolabelled substrate analo-
gues. One approach is to label inhibitors of the enzyme
to which it becomes associated with and disassociate
only very slowly.6 To be useful as a PET tracer, due to
the short half-lives of the radioisotopes (e.g. 11C
t1=2 ¼ 20:4 min), the molecule must be capable of being
rapidly radiolabelled. A recently discovered member of
*Correspondence to: Lutz Schweiger, John Mallard Scottish PET
Centre, School of Medical Sciences, University of Aberdeen, Forester-
hill, Aberdeen AB25 2ZD, UK. E-mail: l.schweiger@biomed.abdn.ac.uk
Contract/grant sponsor: Grampian NHS Endowment grant (Scotland)
any position.14 Here, we report the synthesis of the 11
C
Copyright # 2007 John Wiley & Sons, Ltd.

SYNTHESIS OF (E)-2,30,4,50-TETRAMETHOXY[2-11C]STILBENE 1207
isotope-radiolabelled CYP1B1 inhibitor (E)-2,30,4,50-
sodium borohydride and tetrakis(triphenylphosphane)
palladium(0). The purified product was dissolved in
heptane and reacted with iodine. Column chromato-
graphy provided pure (E)-2-(hydroxy)-30,4,50-tri-
tetramethoxy[2-11C]stilbene.
Results and discussion
methoxystilbene
3
with
a
yield of 14%. The
radiosynthesis of (E)-2,30,4,50-tetramethoxy[2-11C]stil-
bene, 4, is presented in Figure 3. Compound 4 has
been successfully prepared in a single-step reaction by
11C-methylation of 3 using [11C]methyl triflate. For
this, a Pyrex 10 ml test tube was filled under an argon
atmosphere with a solution of 3 in acetonitrile and
potassium carbonate and placed into a dose calibrator
(Capintec CRC-15PET). Once the amount of
To achieve our aim, the 11C radiolabelling of the non-
radioactive precursor (E)-2-(hydroxy)-30,4,50-trimeth-
oxystilbene, 3, the synthetic strategy published by
Kim et al.13 was examined. Kim et al. reported a
synthetic pathway to produce a series of trans stilbene
derivatives and to evaluate their inhibitory activities on
the human cytochrome P450s.13 It was discovered that
the substituent at position 2 of the stilbene skeleton
(Figure 1) plays an imperative role in discriminating
between CYP1B1 and other P450 s CYP inhibitors.13
Despite following the described synthetic method using
the same conditions as Kim et al., the targeted product
3 could not be obtained. Hence, a slightly modified
synthetic route was adopted by ChiroBlock GmbH for
us. The synthetic scheme for the synthesis of (E)-2-
(hydroxy)-30,4,50-trimethoxystilbene, 3, is presented in
Figure 2. 3,5-Dimethoxybenzyltriphenylphosphonium
[
11C]methyl triflate bubbling through the yellow sus-
pension reached a maximum, the reaction mixture was
worked up. Isolation and purification of were
4
accomplished by means of a small disposable C-18
cartridge. After the yellow mixture was removed from
the test tube and introduced to the C-18-based resin,
the cartridge was rinsed with sterile water. The product
4 was eluted with 1.5 ml of ethanol from the cartridge
bromide
1
was synthesized with
a yield of 93%
H3CO
OCH3
Br
according to
a
procedure for the preparation of
O
phosphonium salts.15 For this, 3,5 dimethoxybenzyl-
bromide was dissolved in xylene and charged with
triphenylphosphane. The next step involved the protec-
tion of the hydroxy group of 2-hydroxy-4-methoxy-
benzaldehyde. Experiments to carry out the subse-
quent olefination reaction, and to construct the stil-
bene skeleton, by reacting unprotected 2-hydroxy-4-
methoxy-benzaldehyde13 with 1 were unsuccessful.
Therefore, the facile removable allyl-protecting group
was chosen. 2-Hydroxy-4-methoxy-benzaldehyde was
charged with allylbromide to obtain 2-allyloxy-4-meth-
oxy-benzaldehyde 2 with a yield of 37%. The subse-
quent reaction between 1 and 2 resulted in the
unprotected isomer (E)-2-(hydroxy)-30,4,50-trimethox-
ystilbene 3. For this, 1 was reacted with sodium
hydride and 2. The product isolated was reacted with
OCH3
3,5-Dimethoxy benzyl bromide
OH
2-Hydroxy-4-methoxy-benzaldehyde
K2CO3/NaI
PPh /Xylene
3
OCH3
O
H3CO
+
P
O
-
Br
OCH3
1
2
NaH/THF
THF/NaBH /TPPPd(0)
4
I2/Heptane
OCH3
OCH3
5
4
6
3
1
2
OH
OCH3
6'
1'
2'
H3CO
5'
H3CO
4' 3'
OCH3
3
Figure 1 Chemical structure of (E)-2,30,4,50-tetramethoxys-
tilbene.
Figure 2 Synthesis of (E)-2-(hydroxy)-30,4,50-trimethoxystil-
bene 3.
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 1206–1210
DOI: 10.1002.jlcr

1208 L. SCHWEIGER ET AL.
through a sterile 0.22 Micron Filter (Pall) into a sterile
nitrogen-filled vial containing 9 ml of sterile sodium
chloride, 0.9% w/v, solution. Analytical high-perfor-
mance liquid chromatography (HPLC) showed the
was performed using a Gynkotek HPLC system (P580
pump), a Gynkotek column oven (STHS8S) and a
Gynkotek UV detector (UVD340S) coupled in series
with a BIOSCAN NaI detector (B-FC-3200). The HPLC
system was operated using a Phenomenex Luna C-18
column (250 ꢁ 3.0 mm, particle size: 5 mm). The eluent
used was a mixture of HPLC grade acetonitrile and
0.1 M ammonium formiate solution (50:50). The eluent
product to be >95% radiochemically pure in
a
20 ꢀ 2% decay-corrected radiochemical yield (based
upon [11C]carbon dioxide) and to co-elute with a
sample of non-radioactive TMS at the same retention
time of 20 min (Figure 4). The total synthesis time from
end of bombardment (EOB) was 30 min and the specific
activity of 4 was 1.9 ꢀ 0.6 GBq/mmol at EOB. Recently,
a research group working independently on a similar
synthetic approach reported the synthesis of 11C- and
18F-labelled trans stilbenes. Gao et al.16 described,
among other compounds, the synthesis of (E)-3,30,4,50-
tetramethoxy[3-11C]stilbene and (E)-30,4,50-trimethox-
y[4-11C]stilbene as potential probes for aryl hydrocar-
bon receptor in cancers. However, none of their labelled
stilbene derivatives had a methoxy group at position 2
of the stilbene skeleton and would, therefore, have a
lower affinity for CYP1B1 compared with the 11C-
labelled compound that we have produced.
time [min]
0
5
10
15
20
25
30
Experimental
General
Pure (>98%) TMS was purchased from Cayman. Pure
(>95%) (E)-2-(hydroxy)-30,4,50-trimethoxystilbene was
custom synthesized by ChiroBlock GmbH. All other
reagents were purchased from Sigma-Aldrich and were
used without further purification unless otherwise
noted. All used solvents were purified and degassed
according to standard procedures. FT-IR spectra were
recorded using a Nicolet 205 FT-IR spectrometer. Mass
spectra were recorded on a Shimadzu QP5050 mass
0
5
10
15
20
25
30
1
spectrometer H and 13C NMR data were obtained on a
time [min]
Varian Unity Inova 400 MHz and Bruker 400 MHz
spectrometer at 300 K. Chemical shifts are given in
ppm. Melting point analyses were carried out using a
Gallenkamp melting point apparatus. Analytical HPLC
Figure
4 Top: UV-chromatogram of non-radioactive (E)-
2,30,4,50-tetramethoxystilbene 4. Bottom: Radioactivity chro-
matogram of (E)-2,30,4,50-tetramethoxy[2-11C] stilbene (97%,
20.1 min). Both peaks at 5.1 and 17.6 min are unidentified.
OCH3
OCH3
11CH3OSO2CF3
CH3CN/K2CO3
O11CH3
OH
H3CO
H3CO
OCH3
OCH3
3
4
Figure 3 Radiosynthesis of (E)-2,30,4,50-tetramethoxy[2-11C] stilbene 4.
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 1206–1210
DOI: 10.1002.jlcr

SYNTHESIS OF (E)-2,30,4,50-TETRAMETHOXY[2-11C]STILBENE 1209
was filtered and degassed with Grade A helium (BOC)
before use. The flow rate was set at 1.2 ml/min. The
column oven was set at 408C. [11C]Carbon dioxide,
58C for 30 min. A solution of 2 (1.40 g, 0.01 mol) in
10 ml of anhydrous THF was added dropwise. The
reaction mixture was stirred at r.t. overnight. After
concentration of the mixture, the residue was dissolved
in 50 ml of diethylether and washed with water. The
water phase was collected and extracted with diethy-
lether. The organic phases were combined, dried over
sodium sulfate and concentrated. The obtained yellow
oil was dissolved in 50 ml of THF. Then sodium
borohydride (0.09 g, 3.0 ꢁ 10ꢂ3 mol) and tetrakis(triphe-
nyl phosphane)palladium(0) (1 ꢁ 10ꢂ3 g, 8.7 ꢁ 10ꢂ7 mol)
were added. The yellow solution was stirred at room
temperature overnight. After evaporating the solvent,
the residue was purified via column chromatography
on silica gel (ethyl acetate/petrol ether, 1:2). The
product isolated was dissolved in 50 ml of heptane.
After adding a catalytic amount of iodine (0.01 g,
3.9 ꢁ 10ꢂ6 mol), the purple solution was refluxed for
2 h. Concentration of the solution and subsequent
purification via column chromatography on silica gel
(ethyl acetate/petrol ether, 1:2) provided pure (E)-2-
(hydroxy)- 30,4,50-trimethoxystilbene 3 as an orange
viscous oil (0.27 g, 14%). IR (NaCl, film) cmꢂ1: 3390
(–OH), 3050–3000 (CH arom), 2840 (CH–aliph), 1460,
1590 (C¼¼C), 1290, 1030 (C–O). 1H NMR: (400 MHz,
((CD3)2S¼¼O)): 9.77 (s, OH, 1H), 7.41 (d, J ¼ 9:6 Hz,
1H), 7.24 (d, ¼¼CH–Ph, J ¼ 16:4 Hz, 1H), 6.95 (d,
¼¼CH–Ph, J ¼ 16:4 Hz, 1H), 6.61 (d, J ¼ 2:1 Hz, 2H),
6.38–6.31 (m, 3 H), 3.71 (s, 2 ꢁ –OCH3, 6H), 3.66
(s, –OCH3, 3H). 13C NMR: (100 Hz, ((CD3)2S¼¼O)):
161.3, 160.6, 156.9, 140.8, 128.3, 126.3, 124.8,
117.4, 106.2, 104.5, 101.9, 99.9, 55.8, 55.7. EIMS
m/z: 286 (Mþ, 100%), 137, 255.
[
11C]methyl iodide and [11C]methyl trifluoromethane-
sulfonate radiosyntheses have been carried out accord-
ing to procedures described previously.17
3,5-Dimethoxybenzyltriphenylphosphonium
Bromide (1)
To a stirred solution of 3,5-dimethoxybenzyl bromide
(5.0 g, 0.02 mol) in 40 ml of xylene, a solution of
triphenylphosphane (5.96 g, 0.02 mol) in 10 ml of
xylene was added dropwise. The orange suspension
was refluxed overnight. The flask was cooled down to
r.t. and then placed in the freezer overnight. The
precipitation was filtered, washed with cold xylene
and dried. 3,5 Dimethoxybenzyltriphenylphosphonium
bromide, 1 was obtained as beige crystals (9.95 g,
93%). M.p. 274–2758C (lit.18: m.p. 2758C). 1H NMR:
(400 MHz, CDCl3): 7.74–7.57 (m, 3 ꢁ C6H5, 15 H), 6.29
(d, H-2, H-6, J ¼ 2:4 Hz, 2 H), 6.25 (t, H-4, J ¼ 2:4 Hz,
1H), 5.25 (d, –CH2, J ¼ 14 Hz, 2H), 3.49 (s, 2 ꢁ –OCH3,
6 H).
2-Allyloxy-4-methoxy-benzaldehyde (2)
To A mixture of 2-hydroxy-4-methoxy-benzaldehyde
(3.00 g, 0.02 mol) and allylbromide (2.39 g, 0.02 mol),
potassium carbonate (K2CO3) (4.09 g, 0.03 mol) and a
catalytic amount of sodium iodide (3.0 ꢁ 10ꢂ3 g,
2.0 ꢁ 10ꢂ7 mol) was added. The reaction mixture was
stirred at r.t. overnight. After 12 h, the yellow mixture
was heated up to 408C and stirred for 18 h. Twenty
milliliters of acetone was added to the brown solution
and stirred for an additional 20 h at 408C. After
concentration of the mixture, the residue was dissolved
in dichloromethane. Purification by column chromato-
graphy on silica gel (ethylacetate/petrolether, 1:7)
provided 2-allyloxy-4-methoxy-benzaldehyde 2 as a
white solid (1.4 g, 37%). M.p. 388C (lit.19: m.p. 37.9–
38.18C). 1H NMR: (400 MHz, CDCl3): 10.29 (s, CHO,
1H), 7.75 (d, H-6, J ¼ 8:4 Hz, 1H), 6.48 (dd, H-5,
J ¼ 8:4, 2.4 Hz, 1H), 6.37 (d, H-3, J ¼ 2:4 Hz, 1H),
6.04–5.97 (m, OCH2CH¼¼CH2, 1H), 5.41–5.26 (m,
OCH2CH¼¼CH2, 2H), 4.57–4.55 (m, OCH2CH¼¼
CH2, 2H), 3.80 (s, –OCH3, 3H), (cf. Reference 19).
(E)-2,30,4,50-tetramethoxy[2-11C]stilbene radiosynth-
esis (4)
The synthetic strategy for 4 is based on the metho-
dology previously described for the production of
[N-methyl-11C]methylene blue.17 (E)-2-(hydroxyoxy)-4,
30,50-trimethoxystilbene, 3, (1.0ꢁ 10ꢂ3 g, 3.5 ꢁ 10ꢂ6 mol)
was dissolved in 200 ml of anhydrous acetonitrile in
a
dry Pyrex test tube under argon atmosphere
and treated with potassium carbonate (0.01 g,
7.24 ꢁ 10ꢂ5 mol). The test tube was sealed with a
rubber stopper and heated up to 378C for 2 h. The
methylating agent [11C]methyl triflate was bubbled
through the test tube via a spinal needle (18 GA,
(E)-2-(hydroxy)-30,4,50-trimethoxystilbene (3)
90 mm). After the amount of [
11C]methyl triflate
reached a maximum, the reaction mixture was trans-
ferred onto a C-18 Sep-Pak Plus cartridge (Waters)
which was consequently washed with 15 ml of sterile
water. Then the cartridge was rinsed with 1.5 ml of
ethanol to provide 4. Identification of 4 was carried
Compound 1 (3.60 g, 7.0 ꢁ 10ꢂ3 mol) was added under
nitrogen atmosphere to a cooled suspension of sodium
hydride (0.25 g, 0.01 mol) in 40 ml of anhydrous tetra-
hydrofuran (THF). The red suspension was stirred at
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 1206–1210
DOI: 10.1002.jlcr

1210 L. SCHWEIGER ET AL.
out using analytical HPLC. The synthesis of the
radiolabelled product 4 was confirmed via co-injection
with a commercial sample of TMS. The retention time of
TMS in the UV-chromatogram (300 nm) was identical to
the retention time of 4 in the radioactivity chromato-
gram (Figure 4). In a typical radiochemical experiment,
11 GBq of [11C]carbon dioxide could be converted into
0.74 GBq of 4 within an overall synthesis, purification
and formulation time of 30 min (from EOB). The decay-
corrected radiochemical yield, based on [11C]carbon
dioxide, was 20 ꢀ 5% and the radiochemical purity of
4, measured by analytical HPLC, exceeded 95%. The
specific radioactivity obtained was 1.9 ꢀ 0.6 GBq/mmol
at EOB. Determination of the specific radioactivity of 4
was carried out by means of analytical HPLC according
to Mock et al.20
3. Nunez R, Macapinlac HA, Yeung HWD, Akhurst T,
Cai SD, Osman I, Gonen M, Riedel E, Scher HI,
Larson SM. J Nucl Med 2002; 43: 46–55.
4. Nieweg OE, Pruim J, van Ginkel RJ, Hoekstra HJ,
Paans AMJ, Molenaar WM, Koops HS, Vaalburg W.
J Nucl Med 1996; 37: 257–261.
5. De Potter T, Flamen P, Van Cutsem E, Penninckx F,
Filez L, Bormans G, Maes A, Mortelmans L. Eur J
Nucl Med Mol Imaging 2002; 29: 525–529.
6. Zhang J, McCarthy TJ, Moore WM, Currie MG,
Welch MJ. J Med Chem 1996; 39: 5110–5118.
7. Chun YJ, Kim S, Kim D, Lee SK, Guengerich FP.
Cancer Res 2001; 61: 8164–8170.
8. Murray GI, Melvin WT, Greenlee WF, Burke MD.
Ann Rev Pharmacol Toxicol 2001; 41: 297–316.
9. Murray GI, Taylor MC, McFadyen MC, McKay JA,
Greenlee WF, Burke MD, Melvin WT. Cancer Res
1997; 57: 3026–3031.
Conclusions
10. Larsen MC, Angus WGR, Brake PB, Eltom SE,
Sukow KA, Jefcoate CR. Cancer Res 1998; 58:
2366–2374.
In conclusion, we have synthesized (E)-2,30,4,50-tetra-
methoxy[2-11C]stilbene by means of a 11C-methylation
reaction with a radiochemical purity greater than 95%
in a 20 ꢀ 5% radiochemical yield , based on [11C]car-
bon dioxide, as well as a specific average activity of
1.9 ꢀ 0.6 GBq/mmol at EOB. To our knowledge, (E)-
2,30,4,50-tetramethoxy[2-11C]stilbene is a new com-
pound which may be a potential, universal tumour
in vivo PET tracer.
11. Eltom SE, Larsen MC, Jefcoate CR. Carcinogenesis
1998; 19: 1437–1444.
12. Cheung Y-L, Kerr AC, McFadyen MCE, Melvin WT,
Murray GI. Cancer Lett 1999; 139: 199–205.
13. Kim S, Ko H, Park JE, Jung S, Lee SK, Chu Y-J.
J Med Chem 2002; 45: 160–164.
14. Guengerich FP, Chun YJ, Kim D, Gillam EMJ,
Shimada T. Mutat Res/Fundam Mol Mech Mutagen
2003; 523: 173–182.
Acknowledgements
This work was funded by a Grampian NHS Endowment
grant (Scotland).
15. Shi M, Xu B. J Org Chem 2002; 67: 294–297.
16. Gao M, Wang M, Miller KD, Sledge GW, Hutchins
GD, Zheng QH. Bioorg Med Chem Lett 2006; 16:
5767–5772.
REFERENCES
17. Schweiger L, Craib S, Welch A, Sharp P. J Labell
Compd Radiopharm 2003; 46(13): 1221–1228.
18. Pettit GR, Grealish MP, Jung MK, Hamel E, Pettit
RK, Chapuis JC, Schmidt JM. J Med Chem 2002;
45: 2534–2542.
1. Shreve PD. In Whole-body PET Imaging Methods,
Positron Emission Tomography – Basic Science and
Clinical Practice, Valk PE, Bailey DL, Townsend
DW, Maisey MN (eds). Springer: London, 2004;
481–494.
19. Lee JI, Lee HS, Kim BH. Synth Commun 1996; 26:
3201–3215.
2. Sasaki M, Kuwabara Y, Yoshida T, Nakagawa M,
Fukumura T, Mihara F, Morioka T, Fukui M,
Masuda K. Eur J Nucl Med 1998; 25: 1261–1269.
20. Mock BH, Glick-Wilson BE, Zheng Q-H, De Grado
TR. J Label Compd Radiopharm 2005; 48: S224.
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 1206–1210
DOI: 10.1002.jlcr