New approach for the synthesis of [18F]fluoroethyltyrosine for cancer imaging: Simple, fast, and high yielding automated synthesis

Article abstract of DOI:10.1016/j.bmc.2009.09.029

O-(2-[¹⁸F]fluoroethyl)-L-tyrosine ([¹⁸F]FET) is one of the first ¹⁸F-labeled amino acids for imaging amino acid metabolism in tumors. This tracer overcomes the disadvantages of [ ¹⁸F]fluorodeoxyglucose, [¹⁸

Full text of DOI:10.1016/j.bmc.2009.09.029

Bioorganic & Medicinal Chemistry 17 (2009) 7441–7448
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
New approach for the synthesis of [18F]fluoroethyltyrosine for cancer
imaging: Simple, fast, and high yielding automated synthesis
a
a
a
b
a
M. Zuhayra a, , A. Alfteimi , C. Von Forstner , U. Lützen , B. Meller , E. Henze
*
a Klinik für Nuklearmedizin, Universitätsklinikum Schleswig-Holstein Campus Kiel, Arnold-Heller-Strasse 9, 24105 Kiel, Germany
b Abteilung Nuklearmedizin, KFO179, Georg-August-Universität Goettingen, Robert-Koch-Str. 40, 37075 Goettingen, Germany
a r t i c l e i n f o
a b s t r a c t
O-(2-[18F]fluoroethyl)-
acid metabolism in tumors. This tracer overcomes the disadvantages of
L
-tyrosine ([18F]FET) is one of the first 18F-labeled amino acids for imaging amino
Article history:
Received 7 May 2009
Revised 7 September 2009
Accepted 17 September 2009
Available online 20 September 2009
[
18F]fluorodeoxyglucose,
[
18F]FDG, and 11C]methionine, 11C]MET. Nevertheless, the various synthetic methods providing
[ [
18F[FET] exhibit a big disadvantage concerning the necessity of two purification steps during the synthesis
including HPLC purification, which causes difficulties in the automation, moderate yields, and long synthe-
sis times >60 min.
Keywords:
F-18
Fluoroethyltyrosine
PET
A new approach for the synthesis of [18F]FET is developed starting from 2-bromoethyl triflate as precursor.
After optimization of the synthesis parameters including the distillation step of [18F]-FCH2CH2Br combined
with the final purification of [18F]FET using a simple solid phase extraction instead of an HPLC run the syn-
thesis [18F]FET could be significantly simplified, shortened, and improved. The radiochemical yield (RCY)
was about 45% (not decay corrected and calculated relative to [18F]FÀ activity that was delivered from
the cyclotron). Synthesis time was only 35 min from the end of bombardment (EOB) and the radiochemical
purity was >99% at the end of synthesis (EOS). Thus, this simplified synthesis for [18F]FET offers a very good
option for routine clinical use.
Fluoroethylation
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
One of the first and most successful 18F-labeled amino acids is
O-(2-[18F]fluoroethyl)- -tyrosine ([18F]FET 4).5–10 For synthesizing
L
Radiolabeled amino acids and their analogues are established
tracers in diagnostic oncology.1 Their use in the single-photon
emission tomography (SPECT) and positron emission tomography
(PET) is based on their enhanced accumulation into malignant
transformed cells by increased expression of amino acid transport-
ers. This is assumed to reflect an enhanced amino acid metabolism
and protein synthesis in growing tumors.2
[
18F]FET there are two approaches by nucleophilic substitution
reaction (Scheme 1). The first one is a two-step reaction starting
with the fluorination of 1,2-bis(tosyloxy)ethane (DITOS 1) to form
1-tosyloxy-2-[8F]fluoroethane (2) followed by fluoroethylation of
unprotected L-tyrosine di-sodium salt (Ty-Di-Salt 3) and finally
using HPLC for purification to get 21% [18F]FET 4 in 60 min.5 The
second possibility is the synthesis of [18F]FET via direct nucleo-
philic radiofluorination of the protected precursor O-(2-tosyloxy-
ethyl)-N-trityl-tyrosine tert-butylester (5) yielding about 36%
Currently the most used radiolabeled amino acids in nuclear
medicine are 3-[123I]iodo-
a
-methyl-
-methionine ([11C]MET) for PET.3 Although
11C]MET is the preferably tracer (due to the better imaging proper-
L
-tyrosine ([123I]IMT) for SPECT
and
[
[
11C-methyl]-
L
[
18F]FET in 80 min.6 In both approaches considerable efforts con-
cerning the purification using HPLC are required and consume a
lot of precious time causing long synthesis times and low yields.
Later the introduction of a simple solid phase extraction instead
of HPLC to separate the final product after direct nucleophilic
radiofluorination of the protected precursor N-Boc-(O-2-tosyloxy-
ties of PET technology and its incorporation into proteins) the short
half-life of 11C limited its widespread use to few PET centers with
own cyclotron. For this reason 18F-labeled amino acids were
developed particularly for clinical routine (half-life time of 18F is
110 min). Compared with [18F]fluorodeoxyglucose ([18F]FDG) radio-
labeled amino acids are more specific tracers for the differentiation
between tumor and inflammation since the expression of transport-
ers for amino acids is not stimulated by cytokines.4
ethyl)-L
-tyrosine methyl ester led to [18F]FET 4 in only 35% yield
after 60 min.7
Recently we described a new protocol for fast and high yielding
fully automated synthesis of [18F]fluoroethylcholine via 18F-fluo-
roethylation using 2-bromoethyl triflate (BETfO 6).11 Now we used
also BETfO 6 as starting material for synthesizing [18F]FET too and
could provide after optimization of the synthesis parameters a no-
vel and very simple synthesis approach to synthesis [18F]FET ready
* Corresponding author. Tel.: +49 4315973101; fax: + 49 4315973065.
E-mail address: mzuhayra@nuc-med.uni-kiel.de (M. Zuhayra).
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.09.029

7442
M. Zuhayra et al. / Bioorg. Med. Chem. 17 (2009) 7441–7448
[
18F]BFE 7 with Ty-Di-Salt 3 can be easily performed under pure con-
ditions in dimethylsulfoxide (DMSO) resulting in higher specific
activity for [18F]FET 4.
The individual reaction steps shown in Scheme 2, the first reac-
tion (formation of [18F]BFE 7), the distillation of [18F]BFE 7 and the
second reaction (formation of [18F]FET 4), were subsequently and
systematically studied and optimized.
2.1. First reaction: formation of [18F]BFE 7
The adequate formation of the activated [18F]fluoride either as
[
18F]KF–K222-complex or as [18F]Bu4NF by using K2CO3/KryptofixÒ
or tetrabutylammonium hydrogencarbonate (TBA) is essential for
the subsequent fluorination reaction of BETfO 6 leading to the vol-
atile intermediate [18F]BFE 7.
As described in our previous work it was found for the case of
using of KF–K222-complex that the maximum yield of [18F]BFE 7
was achieved by using 80 lmol K2CO3/K222 and 95 lmol BETfO 6
as well as helium flow rate and drying temperature of 100 mL/
min and 90 °C, respectively (Fig. 1).11 TBA exhibits lower basic
properties compared with KryptofixÒ. Therefore higher radiochem-
ical yield of [18F]BFE 7 would be expected due to less elimination
by-products. But the results in Figure 1 show no significant differ-
ences concerning formation of [18F]BFE 7 by using [18F]Bu4NF in-
stead of [18F]KF–K222-complex. The data indicate an optimum for
the formation of [18F]BFE 7 by using [18F]Bu4NF at 100 mL/min
for the helium flow rate and 110 °C for the drying temperature.
These results are summarized in Figure 1. In both cases no transfer
of TBA or KryptofixÒ from rector one into reactor two during the
distillation step was detected, due to their high vapor pressures.
2.2. Distillation of [18F]BFE 7
The distillation of [18F]BFE 7 (bp 71.5 °C) into the second reactor
was started directly after the drying step by adding 95 lmol BETfO
6 that was dissolved in 0.5 mL o-DCB to the activated [18F]fluoride.
This simple distillation at 100 °C under helium flow transfers the
intermediate [18F]BFE 7 in almost pure form into the second reac-
tor containing Ty-Di-Salt 3 and sodium iodide (NaI) in DMSO. As
shown in the distillation profile presented in Figure 2 the most
radioactivity (83%) is already transported after 6 min from reactor
one into reactor two. About 10% of the start activity remains in
reactor one while about 7% of the start activity is lost by the
distillation.
Scheme 1. Common syntheses of [18F]FET 4.
for human use within 35 min in high radiochemical yield (RCY) of
45% and without the need of any HPLC run.
2. Results and discussion
BETfO 6 was subjected to nucleophilic substitution with [18F]KF–
KryptofixÒ-complex ([18F]KF–K222) as well as with [18F]tetrabutyl-
ammonium fluoride ([18F]Bu4NF) to form the volatile 1-bromo-2-
The dependence of the portion of the intermediate [18F]BFE 7
transferred through the distillation on the temperature and the he-
lium flow rate by using both [18F]KF–K222, and [18F]Bu4NF is shown
in Figure 3. It turned out that the maximum yield of transferred
radioactivity was obtained when a flow rate of 100 mL/min and a
temperature of 100 °C were applied. Higher temperatures led to
less transferred material due to decomposition of reactants. Gas
chromatographic analysis of the distilled material showed no pres-
[
18F]fluoroethane ([18F]BFE 7) in 1,2-dichlorobenzene (o-DCB) (first
reaction in Scheme 2). Because of the significantly lower boiling
point of [18F]BFE 7 (bp 71.5 °C)12 compared with the precursor BETfO
6 (bp 230 °C) and the solvent, o-DCB: (bp 179 °C), the intermediate
[
18F]BFE 7 could be elegantly distilled in a simple way into a second
reactor within 8 min at 100 °C leaving all impurities of the first reac-
tion behind in the first reactor. In this way, the second reaction of
[
18F]KF-K222; [18F]Bu4NF
18F
OTrf
first reaction
Br
Br
Br
DMSO,
o
-DCB
bp 189 , 180 °C
6
7
bp 230 °C
bp 71.5 °C
distillation
C
°
0
0
1
O
18F
NH2
O
Ty-Di-Salt 3
18F
second reaction
OH
DMSO, NaI
7
4
Scheme 2. Two-pot synthesis of [18F]FET 4.

M. Zuhayra et al. / Bioorg. Med. Chem. 17 (2009) 7441–7448
7443
First reaction
Kryptofix/carbonate 80 µmol TBA/Carbonate 80 µmol
Drying temperature 100 °C
He flow 100 mL/min
BETfO 80 µmol
Kryptofix/carbonat 80 µmol
Drying temperaure 90 °C
He flow 100 mL/min
Kryptofix/carbonate 80 µmol
Drying temperatue 90 °C
BETfO 95 µmol
100%
80%
60%
40%
20%
0%
He flow 100 mL/min
BETfO 95 µmol
He flow 100 mL/min
BETfO 95 µmol
110
100
95
80
90
95
100
90
85
110
100
115
120
70
60
100
60
55
85
90
40
30
40
20
80
Kryptofix/Carbonate [µmol]
BETfO [µmol]
Helium flow rate [mL/min]
Drying temperature for
Kryptofix/Carbonate [°C]
Drying temperature for
TBA/Carbonate [°C]
Figure 1. Dependence of the radiochemical yield of [18F]BFE 7 on the amount of reactants, helium flow and drying temperature by using [18F]KF–K222, and [18F]Bu4NF (n = 3).
100%
Sum
90%
80%
Reactor 2
70%
Time 8 min
He flow 100 mL/min
BETfO 80 µmol
60%
Kryptofix/carbonat 80 µmol
50%
40%
30%
20%
10%
0%
Drying temperaure 90 °C
Reactor 1
0
1
2
3
4
5
6
7
8
Time [min]
Figure 2. Distillation profile of [18F]BFE 7 at 100 °C.
ence of BETfO 6 (bp 230 °C) in the second reactor when the temper-
ature was kept ꢀ130 °C (detection limit 0.0014 mg/mL).
to optimize the radiochemical yield of [18F]FET 4 the effect of reac-
tion time and reaction temperature as well as the effect of the
amount of Ty-Di-Salt 3 were subsequently studied. As it was
known that NaI increases the yield of fluoroethylation13,14 we stud-
ied also the effect of NaI on the second reaction. As shown in Figure
4 the optimization of the reaction time and reaction temperature
both for the reaction with NaI as well as for the reaction without
NaI resulted in a maximum yield obtained at 110 °C and 15 min,
Applying the optimized conditions for the first reaction and for
the distillation step has enabled to obtain the intermediate
[
18F]BFE 7 in a pure form with a RCY of 73% in only 15 min without
using HPLC or solid phase extraction (SPE). This is a significant
improvement against the direct labeling approach using DITOS
1,5 which provides the corresponding intermediate [18F]-fluoroeth-
yltosylate ([18F]FETos) in a RCY of only 65% by using both HPLC-run
and SPE for purification (see Scheme 1).
respectively. By adding of 33 lmol (5.5 mg) NaI the radiochemical
yield of [18F]FET 4 could be increased by 30% compared to the reac-
tion without NaI.
2.3. Second reaction: formation of [18F]FET 4
The study of the effect of the amount of tyrosine on the radio-
chemical yield of [18F]FET 4 was started by the use of various quan-
The second reaction between [18F]BFE 7 and freshly prepared
Ty-Di-Salt 3 was carried out after distilling [18F]BFE 7 in a second
tities of tyrosine which were always dissolved in 28 lL of 25%
sodium methanolate solution (CH3ONa) in methanol. Figure 4
reactor containing Ty-Di-Salt 3, NaI and 800
l
L DMSO. In order
shows that the radiochemical yield of [18F]FET 4 increases by

7444
M. Zuhayra et al. / Bioorg. Med. Chem. 17 (2009) 7441–7448
Distillation
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Time 8 min
He flow 100 mL/min
BETfO 80 µmol
Kryptofix/carbonat 80 µmol
Drying temperaure 90 °C
Time 8 min
He flow 100 mL/min
BETfO 80 µmol
TBA/carbonat 80 µmol
Drying temperaure 90 °C
Time 8 min
BETfO 80 µmol
Kryptofix/carbonat 80 µmol
Drying temperaure 90 °C
Distillation temperature 130 °C
130
100
110
140
130
120
120
90
140
110
100
60
100
85
30
80
Distillation temperature by using Kryptofix [°C]
Distillation temperature by using TBA [°C]
Helium [mL/min]
Figure 3. Dependence of the yield of distillation of [18F]BFE 7 on temperature and helium flow (n = 3).
increasing the amount of tyrosine, but starts to fall after reaching a
maximum by 63.4 mol. For further investigations, we then varied
the amount of the added sodium methanolate but kept the quan-
tity of tyrosine constant by 63.4 mol. We found out that when a
mass spectra of compound 8 and [
19F]fluoroethyltyrosine
l
([19F]FET 9) taken with both cone voltages at 20 and 40 V are dem-
onstrated in Figure 6. Both compound 8 and [19F]FET 9 have the
same highest peak at M+1 = 228.1 (C11H15FNO3)+ (spectra A and
C) but they differ in their fragmentation musters at higher cone
voltage of 40 V (spectra B and D). With the help of the fragment
mass at m/z = 136.0 which belongs to the fragment (C8H10NO)+
(10) that only appears in the spectrum of compound 8 at 40 V
(spectrum D) we could identify compound 8 to be 2-fluoroethyl es-
ter of tyrosine because the fragment 10 is built by breaking the C–C
l
molar ratio of CH3ONa over tyrosine of less than two was used
the radiochemical yield of [18F]FET 4 decreased and the HPLC radio
chromatogram of the reaction solution exhibited in addition to
[
18F]FET
4 a second radioactive substance, compound 8, by
13.5 min, which is assumed to be the 2-fluoroethyl ester of tyro-
sine (Fig. 5b and c). This assumption corresponds to the fact that
a double molar excess of CH3ONa over tyrosine is necessary for
the deprotonation of the phenolic group and building of Ty-Di-Salt
3. For the unambiguous identification of compound 8 we per-
bond between the
a and b carbon atoms from the tyrosine ester. It
is unlikely that the same fragment is derived from [19F]FET 9, since
it has to be built by double fragmentation. On the other hand
formed the cold reaction with [19F]KF with 126.9
and 28 L of 25% sodium methanolate solution in 800
l
mol tyrosine
L DMSO
breaking the same C–C bond between the a and b carbon atoms
l
l
in [19F]FET 9 leads to the fragment (C10H13FNO)+ (11) with m/
z = 182.1 which is only present in the spectrum of [19F]FET 9 (spec-
trum B).
(CH3ONa molar ratio = 1.0). The HPLC fraction by 13.5 min was iso-
lated and investigated with ESI mass spectrometry. The resulting
Second reaction
100%
Temperature 100 °C
Ty-Di-Sa 64 µmol
Time 15 min
NaI 33 µmol
Ty-Di-Sa 64 µmol
25 % Methanolic CH3ONa 28 µL
Time 15 min
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
25 % Methanolic CH3ONa 28 µL
Temperature 100 °C
NaI 0 µmol
Time 15 min
Temperature 100 °C
NaI 0 µmol
Ty-Di-Sa 50 µmol
110
100
25 % Methanolic CH3ONa 28 µL
33
120
25 % Methanolic CH3ONa 28 µL
43
80
15
64
83
NaI 0 µmol
Ty-Di-Sa 64 µmol
0
25 % Methanolic CH3ONa 28 µL
Time 15 min
20
15
50
110
10
5
120
40
100
90
28
80
Reaction time [min]
Reaction temperature
without NaI [°C]
Reaction temperature with
33 µmol NaI [°C]
Ammount of Ty-Di-Sa 3
Ammount of NaI [µmol]
Figure 4. Dependence of the radiochemical yield of [18F]FET 4 on reaction time, reaction temperature as well as on the amount of tyrosine and NaI (n = 3).

M. Zuhayra et al. / Bioorg. Med. Chem. 17 (2009) 7441–7448
7445
Figure 5. Radio HPLC chromatograms of reaction solutions of reaction two by different molar excess of NaOH over tyrosine.
The effect of the amount of CH3ONa on the radiochemical yield
of [18F]FET 4 is dramatic, because of the necessity to deprotonate
the phenolic hydroxyl group of tyrosine before it can act as a nucle-
ophilic agent toward [18F]BFE 7. Due to the lower pKs value of the
carboxyl group of tyrosine compared with the pKs value of the
phenolic hydroxyl group (pKsCOOH = 2.2; pKsOH = 10.07)15 a molar
ratio of CH3ONa/tyrosine ꢁ2 is necessary to completely deproto-
nate the phenolic hydroxyl group and avoid the formation of the
2-fluoroethyl ester of tyrosine.
thane 2 with RCY about 30%,5 this is an improvement of more than
35%. Also, the total RCY of the here described approach with 45% is
about 23% higher than the total RCY of the indirect labeling ap-
proach and about 10% more than the total RCY provided by direct
labeling methods using protected precursors.6,7 In addition the
presented approach allows the synthesis of [18F]FET 4 in signifi-
cantly higher specific activity without need of purification with
the HPLC and reduces the synthesis time by half.
Due to the distillation step, which allows to obtain very pure
The RCY of the second reaction after optimization of all reaction
parameters amounts to 65%. Compared with the second reaction of
the indirect labeling approach using 1-tosyloxy-2-[18F]fluoroe-
[
18F]BFE 7, we could achieve high specific activity of more than
80 GBq/
proach leads to 18.5 GBq/
l
mol for [18F]FET 4. In comparison, the direct labeling ap-
mol.6
l
Figure 6. ESI mass spectra of compounds 8 and 9 at cone voltages 20 and 40 V.

7446
M. Zuhayra et al. / Bioorg. Med. Chem. 17 (2009) 7441–7448
2.4. Chemical and radiochemical purity
two upon distillation at 100 °C . Thus, they do not interfere with
the second reaction. The only compounds that might possibly be
also transferred into reactor two would be o-DCB (bp 180 °C) and
BETfO 6 (bp 230 °C) due to their considerable vapor pressure. GC
measurements of the distillate, however, clearly revealed that only
All non-volatile reactants and products of the first reaction,
K222, K2CO3, [18F]KF–K222 as well as TBA and [18F]Bu4NF stay be-
hind in reactor one and only [18F]BFE 7 is transferred into reactor
a trace amount of o-DCB (5 lL/mL) is actually present in the distil-
late whereas BETfO 6 was not detected at all (detection limit
0.0014 mg/mL). This is an important finding because BETfO 6 could
also react with Ty-Di-Salt 3 to give by-products which would cause
problems in the following separation of [18F]FET 4 by solid phase
extraction (SPE). The contamination with o-DCB, however, could
easily be removed together with DMSO by diluting the reaction
mixture with 0.1 M ammonium acetate buffer (NH4Ac), loading
Table 1
Chemical purity of the final ready for humane use solution of [18F]FECh 4 in saline
detected with the gas chromatography
Solvent
L/mL)
Det. limit (
Acetone
Ethanol
o-DCB
DMSO
Acetonitrile
(l
1.19
100
—
—
—
lL/mL)
0.005
0.006
0.002
0.007
0.020
Figure 7. Chemical and radiochemical purity detection with (A) gas chromatography, (B) radio HPLC with diode array detection.

M. Zuhayra et al. / Bioorg. Med. Chem. 17 (2009) 7441–7448
7447
onto the C18 cartridges, and subsequent cleaning with ammonium
acetate buffer and 5%V ethanol containing saline. All educts were
hereby eluted into the waste. [18F]FET 4 was finally extracted from
the C18 cartridges with 10 mL 10%V ethanol containing saline. As
shown in Figure 7, the investigation of final sterile solution of
synthesis EOS. Gas chromatography was performed with Varian
3350 spectrometer and FID detection.
3.2. 2-Bromoethyl triflate BETfO 6
[
18F]FET 4 via gas chromatography showed only the present of
BETfO 5 was synthesized as described in the literature.16 2-Bro-
moethanol (1 g, 8.0 mmol) was dissolved in 2.05 mL of 2,6-lutidine
(17.6 mmol), diluted with 10 mL of CH2Cl2 and cooled to 0 °C. Tri-
fluoromethanesulfonic acid anhydride (2.83 mL, 16.8 mmol) was
added dropwise. After being stirred for 30 min, the solvent was re-
moved in vacuum and the product was distilled (50 °C at 0.5 mm
Hg) to give 1.1 g (54%) of BETfO 6 as a colorless liquid: GC chro-
matogram, single peak, tR = 7.82 min (column RtxÒ 200, 0.25 mm
ID, 30 m, carrier gas N2, 0.95 mL/min, injector 250 °C, program
50 °C (5 min hold, than 10 °C/min)).
minimal amounts of acetone from the cleaning procedure
(1.19 L/mL) and ethanol (10%V). No other solvents could be de-
l
tected (Table 1). Figure 7 shows HPLC radio chromatogram, GC
chromatogram and ultra-violet spectrum of the final product.
According to the high vapor pressures of TBA and KryptofixÒ,
which prevent their transfer during the distillation step into reac-
tor two, the investigation of the final sterile solution of [18F]FET 4
via LC–MS showed no presence of both compounds.
The purity and identity were checked by RP C18 chromatogra-
phy (Nucleosil 100-5 HD 4–250 mm, solvents: A: NH4Ac (0.1 M);
B: acetonitrile; gradient: 0–5 min 100% A, 5–30 min 0–100% B;
flow rate: 1.5 mL/min). Typical chromatograms of the quality con-
trol are shown in Figure 7.
3.3. Radiosynthesis
The synthesis of [18F]FET 4 was carried out in the style of GMP.
[
18F]FET 4 was produced in a hot cell equipped with a commer-
cially available automated synthesis module using sterile filtered
helium as gas carrier. A schematic diagram of the automated syn-
thesis module is presented in Figure 8. All starting materials were
tested using validated methods and qualified equipment. The qual-
ity control involved measurement of the chemical and the radio-
chemical purity via HPLC, gas chromatography and the pH and
other parameters were tested.
3. Experimental
3.1. Materials and methods
Chemicals were purchased from commercial sources and were
used without further purification. The Chromafix C18 cartridges
were purchased from Macherey Nagel, [18F]fluoride was produced
via the 18O(p,n)18F reaction with a CTI RDS 111 cyclotron (Berlin).
HPLC was performed with HP 1100 pump with UV and gamma
radiation detection (Gabi, Raytest), ESI spectra were performed
with Platform LC (Waters). The radiochemical yield was not decay
corrected and calculated relative to [18F]FÀ activity that was deliv-
ered from the cyclotron. Synthesis time was measured from the
end of bombardment (EOB) and the specific activity at the end of
[
18F]fluoride ion (2–10 GBq) obtained from the target was
trapped in a small anion exchange column (30 mg, HCO3 form).
The radioactivity was eluted with an aqueous solution of K2CO3
(11 mg; 0.08 mmol) in 0.8 mL of H2O into reactor one. A solution
of K222 (30 mg; 0.08 mmol) in 1.5 mL of CH3CN was added, and
the mixture was evaporated at 90 °C under reduced pressure with
a helium flow of 100 mL/min for 2 min and without helium flow
for additional 2 min. 0.5 mL of o-DCB were then added to the
Figure 8. Schematic diagram of the automated synthesis module.

7448
M. Zuhayra et al. / Bioorg. Med. Chem. 17 (2009) 7441–7448
dried [18F]KF–K222. After 1 min of stirring at 100 °C BETfO 5
(15
L) dissolved in 0.5 mL of o-DCB was added while [18F]BFE
7 was transferred within 8 min into the second reactor containing
11.5 mg tyrosine (63.4 mol), 28 L of 25% sodium methanolate
solution (CH3ONa) in methanol, 5 mg NaI (33 mol), and 800
Supplementary data
l
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2009.09.029.
l
l
l
lL
of DMSO. The second reactor was then tempered for 15 min at
110 °C. Afterwards the temperature was decreased to 60 °C,
20 mL of NH4Ac (0.1 mol/L) were added and the reaction mixture
was passed through two C18 Chromafix cartridges (Macherey Na-
gel). The cartridges were then washed with 10 mL of NH4Ac and
3 mL of 5%V ethanol containing saline, respectively. The [18F]FET
4 was then eluted with 10 mL of 10%V ethanol containing saline.
If necessary the [18F]FET 4 solution can be diluted with saline to
reduce the EtOH concentration to be, for example, 5%V. After ster-
ile filtration 0.9–4.5 5% GBq [18F]FET 4 were obtained (45 5%).
The radiochemical purity was >99% and the specific activity
References and notes
1. Vaalburg, W.; Coenen, H. H.; Crouzel, C.; Elsinga, P. H.; Långström, B.; Lemaire,
C. Nucl. Med. Biol. 1992, 19, 227.
2. Langena, K. J.; Hamacher, K.; Weckesser, M.; Floeth, F.; Stoffelsa, G.; Bauer, D.;
Coenen, H. H.; Pauleit, D. Nucl. Med. Biol. 2006, 33, 287.
3. Jager, P. L.; Vaalburg, W.; Pruim, J.; de Vries, E. G.; Langen, K. J.; Piers, D. A. J.
Nucl. Med. 2001, 42, 432.
4. Del Sole, A.; Falini, A.; Ravasi, L.; Ottobrini, L.; De Marchis, D.; Bombardieri, E.;
Lucignani Eur. J. Nucl. Med. 2001, 28, 1851.
5. Wester, H. J.; Herz, M.; Weber, W.; Heiss, P.; Senekowitsch-Schmidtke, R.;
Schwaiger, M.; Stöcklin, G. J. Nucl. Med. 1999, 40, 205.
6. Hamacher, K.; Coenen, H. H. Appl. Radiat. Isot. 2002, 57, 853.
7. Wang, H.-E.; Wua, S.-Y.; Chang, C.-W.; Liu, R.-S.; Hwang, L.-C.; Lee, T.-W.; Chen,
J.-C.; Hwan, J.-J. Nucl. Med. Biol. 2005, 32, 367.
8. Krasikova, R. N.; Kuznetsova, O. F.; Fedorova, O. S.; Maleev, V. I.; Saveleva, T. F.;
Belokon, Y. N. Bioorg. Med. Chem. 2008, 16, 4994.
>80 GBq/lmol at the end of synthesis (EOS). Synthesis time was
35 min from EOB.
9. Tang, G.; Tang, X.; Wang, M.; Luo, L.; Gan, M.; Huang, Z. J. Labelled Compd.
Radiopharm. 2003, 46, 661.
10. Wei, W. M.; Zhi, Y. D.; Qiang, L. S.; Xian, W. Y. Sci. China, Ser. B 2007, 50, 276.
11. Zuhayra, M.; Alfteimi, A.; Papp, L.; Lützen, U.; Lützen, A.; Von Forstner, C.;
Meller, B.; Henze, E. Bioorg. Med. Chem. 2008, 16, 9121.
4. Conclusion
The synthesis of O-(2-[18F]fluoroethyl)-
L-tyrosine could be
12. Pattison, F. L. M.; Peters, D. A. V.; Dean, F. H. Can J. Chem. 1965, 43, 1689.
13. Zhang, M.-R.; Furutsuka, K.; Yoshida, Y.; Suzuki, K. J. Labelled Compd.
Radiopharm. 2003, 46, 587.
14. Bauman, A.; Piel, M.; Schirrmacher, R.; Rösch, F. Tetrahedron Lett. 2003, 44,
9165.
15. Jakubke, H.-D.; Jeschkeit, H. Aminosäuren, Peptide, Proteine, Verlag Chemie:
Weinheim, 1982; ISBN 3-527-25892-2, p 40.
16. Chi, D. Y.; Kilbourn, M. R.; Katzenellenbogen, J. A.; Welch, M. J. J. Org. Chem.
1987, 52, 658.
significantly simplified, shortened, and improved. The use of the
non-volatile 2-bromoethyl triflate as the starting material and
the introduction of a distillation step substitute the HPLC purifica-
tion steps in common synthesis procedures making the synthesis
simple and short. At the same time both the radiochemical yield
and the chemical purity could be increased dramatically. Thus,
the routine synthesis of O-(2-[18F]fluoroethyl)-
L-tyrosine became
more practicable and economically profitable.