Direct radiofluorination of [18F]MH.MZ for 5-HT2A receptor molecular imaging with PET

Article abstract of DOI:10.1002/jlcr.2947

Imaging the serotonin 2A neuroreceptor with positron emission tomography has been carried out with [¹¹C]MDL 100907 and [¹⁸F] altanserin for years. Recently, the MDL 100907 analogue [¹⁸F]MH.MZ was developed by combining the

Full text of DOI:10.1002/jlcr.2947

Research Article
Received 23 May 2012,
Revised 19 June 2012,
Accepted 29 June 2012
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/jlcr.2947
Direct radiofluorination of [18F]MH.MZ for 5-
HT2A receptor molecular imaging with PET
a
b
c
b
*
Matthias M. Herth, Vasko Kramer, Nic Gillings, Frank Rösch, and
Gitte M. Knudsena
Imaging the serotonin 2A neuroreceptor with positron emission tomography has been carried out with [11C]MDL 100907
and [18F]altanserin for years. Recently, the MDL 100907 analogue [18F]MH.MZ was developed by combining the increased
selectivity profile of MDL 100907 and the favourable radiophysical properties of fluorine-18. Here, we want to report the
synthesis of [18F]MH.MZ via direct radiofluorination. Unfortunately, the direct radiofluorination did not have any significant
benefits over the indirect labelling method. This is mainly because the precursor for the direct labelling approach is not
completely stable and slowly decomposes. However, only one HPLC separation is necessary for the direct 18F-nucleophilic
labelling procedure, and accordingly, automation is easier.
Keywords: MDL 100907; [18F]MH.MZ; PET; 5-HT2A receptor
analogue of MDL 100907, [18F]MH.MZ (1) and its R-enantiomer
Introduction
(2) (Figure 1) and reported its subsequent evaluation in rodents.8
Serotonin (5-hydroxytryptamine, 5-HT) 2A receptors are of
In vitro and in vivo studies demonstrated that in rodents, this
significant clinical interest because of their involvement in the
radioligand combines the high selectivity of MDL 100907
(Table 1) with the superior isotopic properties of [18F]fluorine.8–11
pathophysiology of human diseases such as depression, Alzheimer’s
disease and schizophrenia.1 In particular, therapeutic effects of
The aim of this study was to simplify the two-step labelling
procedure of [18F]MH.MZ to a direct radiofluorination approach.
atypical antipsychotics may be attributed to antagonistic effects on
these receptors.2 Stimulation of the 5-HT2A receptor with
recreational hallucinogens such as lysergic acid diethylamide elicits
hallucinogenic effects.3
Results
Positron emission tomography (PET) is widely used as an
Organic chemistry
advanced tool to image cerebral receptors in vivo, where the
pharmacological parameters of ligand–neuroreceptor interac-
tions can be quantified. Therefore, in vivo PET studies of cerebral
5-HT2A receptors and occupancy by therapeutic drugs, for
example antipsychotics, have provided a significant advance in
the understanding of these disorders.
The reference compound MH.MZ was synthesised as previously
described.10 The precursor (5) was synthesised in a four-step
reaction sequence starting with a condensation reaction between
(3-((tert-butyldiphenylsilyl)oxy)-2-methoxyphenyl)(piperidin-4-yl)
methanol (1) and p-fluorophenylethylbromide in a yield of
approximately 75%. The resulting intermediate (2) was acylated
(yield: 99%) and the tert-butyldiphenylsilyl (TBDPS)-protecting
group removed by NH4F (yield: 76%). Finally, ethyleneditosylate
was reacted with the phenolic moiety of (4) via a Williamson
ether synthesis to give (5) in a yield of 46% (Scheme 1).
Currently, the 5-HT2A receptor antagonist radiotracers [18F]
altanserin (Table 1) and to a lesser extent [11C]MDL 100907 are
used as PET tracers to probe the 5-HT2A receptors in vivo.4 They
have high affinity and selectivity for the 5-HT2A receptor (Table 1)
and generate good target-to-background ratios in humans.
However, both PET radiotracers have limitations; [11C]MDL
100907 displays slow kinetics,5 whereas in vitro binding of [18F]
altanserin lacks 5-HT2A receptor specificity (Table 1) and it gives
rise to lipophilic metabolites that cross the blood–brain barrier.6
Nevertheless, [18F]altanserin is the most commonly used recep-
tor ligand for 5-HT2A receptor imaging. For the reasons given
earlier, it would, however, be advantageous to have access to
an [18F]MDL 100907 derivative that did not generate lipophilic
metabolites.
aCenter for Integrated Molecular Brain Imaging, Copenhagen University Hospital,
Rigshospitalet, Blegdamsvej 9, DK-2100 Copenhagen, Denmark
bInstitute of Nuclear Chemistry, Johannes Gutenberg University-Mainz, Fritz-
Strassmann-Weg 2, 55128 Mainz, Germany
cPET and Cyclotron Unit, Copenhagen University Hospital, Rigshospitalet,
Blegdamsvej 9, DK-2100 Copenhagen, Denmark
In an attempt to meet those criteria, [18F]MDL 100907 has
been synthezised.7 However, because of the very low radioche-
mical yield (RCY) of 2%, this radiotracer is not suitable for clinical
studies. We later reported a two-step synthesis of an 18F-labelled
*Correspondence to: Matthias M. Herth, Center for Integrated Molecular Brain
Imaging, Copenhagen University Hospital, Rigshospitalet, Blegdamsvej 9,
DK-2100 Copenhagen, Denmark.
E-mail: matthias@nru.dk
J. Label Compd. Radiopharm 2012
Copyright © 2012 John Wiley & Sons, Ltd.

M. M. Herth et al.
Table 1. Dissociation constants (Ki (nM)) and chemical structures of altanserin, MDL 100907, (R,S)-MH.MZ and (R)-MH.MZ. Affi-
nities for a large range of receptors are previously published10,12
5-HT2A
0.13
D2/5-HT2A ratio
~475
a1/5-HT2A ratio
Altanserin
~35
MDL 100907
0.36
9.02
0.72
~6250
> 1100
~3730
~350
(R,S)-MH.MZ
~35
(R)-MH.MZ
~465
4-yl)(2-methoxy-3-(2-(tosyloxy)ethoxy)phenyl)methyl acetate (5)
followed by Zemplen deprotection (Scheme 2). The whole
procedure was carried out as a one-pot reaction.
The effect of precursor concentration and reaction solvent on
radioactive labelling kinetics was evaluated (Scheme 2). Reaction
progress was measured with radio–thin layer chromatography
(TLC), and results were verified with analytical HPLC. The
optimised reaction conditions (3 mg precursor, 80 ꢀC, 20-min
reaction time) gave RCYs of approximately 30%.
The effect of varying temperature was not measured because
of a decomposing precursor at 80 ꢀC in dimethyl sulfoxide
(DMSO) and dimethylformamide (DMF) (data not shown).
Increasing the amount of precursor (3–9 mg) did not significantly
change the RCY (data not shown). Finally, Zemplen deacylation
of [18F](Z) using 0.05 NaOMe at 80 ꢀC for 3 min leads to quantita-
tive deprotection resulting in [18F]MH.MZ.
HPLC purification was carried out as described in the experi-
mental section. The radiochromatogram of the crude product
showed a difference in the two retention times of approximately
4 min between the two radioactive compounds, eluting first [18F]
fluoride (~2 min) and then [18F]MH.MZ (~6 min). Ultraviolet (UV)
absorption at 254 nm indicated that the precursor (5) had a
retention time of approximately 12 min (Figure 1).
Figure 1. Analytical HPLC diagram of [ [
18F]MH.MZ, 18F](5), MH.MZ and its
precursor.
Radiochemistry
Radiolabelling of [18F]MH.MZ was carried out by nucleophilic
substitution of [18F]fluoride on (1-(4-fluorophenethyl)piperidin-
www.jlcr.org
Copyright © 2012 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2012

M. M. Herth et al.
Scheme 1. Reagents and conditions: (a) NaI, p-fluorophenylethylbromide, NaHCO3, dimethylformamide, 85 ꢀC, 2 h, 70%; (b) Ac2O, THF, RT, 99%; (c) NH4F, MeOH, reflux,
76%; (d) Cs2CO3, ethyleneditosylate, MeCN, reflux, 16 h, 46%.
In addition, optimization experiments via microwave (MW)
heating were carried out. All experiments were carried out in
MeCN, and radioactive labelling kinetics were evaluated and
optimised on MW heating mode, temperature and time. MW
heating reduced the reaction time to 5 min, whereas RCY was
unchanged (Figure 2).
Discussion
We here report a novel approach to synthesise [18F]MH.MZ in a
one-step radiochemical approach followed by a Zemplen depro-
tection reaction in sufficient yield for imaging studies; the best
conditions of the nucleophilic substitution included MeCN at
80 ꢀC. MW irradiation has been successfully applied in organic
chemistry and radiolabelling. Because of accelerations, higher
yields under milder reaction conditions and higher product
purities have been reported,13,14 we tested MW heating as a
method to increase RCY for [18F]MH.MZ radiolabelling. This
optimised reaction resulted in an RCY of approximately 25%,
only slightly lower than that of a conventional heating system,
but the reaction time could be reduced by 25 min. No radiolysis
was observed for [18F]MH.MZ, but the precursor was sensitive to
air, and unless stored at inert conditions, the RCYs dropped. This
problem could not be overcome by using a higher precursor
Scheme 2. (A) Nucleophilic 18F-labelling strategy to synthesise [18F]MH.MZ and
(B) 18F-labelling kinetics of the intermediate ([18F](5)).
The radiosynthesis including HPLC, SepPakW (Waters, Hedehusene
Denmark) purification and formulation gave a final injectable
solution of [18F]MH.MZ (radiochemical purity > 96%) within
100 min. Specific activities (As) were in average 19.8 GBq/mmol
(range 0.69–65.9 GBq/mmol). Approximately 25 GBq of [18F]
fluorine was used as starting radioactivity, whereby [18F]MH.
MZ was obtained as an injectable solution in an overall RCY
Figure 2. 18F-Labelling kinetics of the intermediate ([18F](5)).
around 15–20%.
J. Label Compd. Radiopharm 2012
Copyright © 2012 John Wiley & Sons, Ltd.
www.jlcr.org

M. M. Herth et al.
concentration. Compared with the two-step labelling procedure solution was quenched with H2O (30 mL), extracted with EtOAc
of [18F]MH.MZ via [18F]FETos7, only one HPLC separation is (3 Â 40 mL), dried (Na2SO4) and evaporated, yielding (3) as a
1
necessary for the direct 18F-nucleophilic labelling procedure, colourless oil (1.4 g, 2.18 mmol, 99%). H NMR (300 MHz, CDCl3)
and accordingly, automation would be easier. However, the d [ppm] = 7.665–7.753 (4H, m), 7.268–7.428 (6H, m), 7.118–7.165
two-step procedure takes roughly the same time, has similar RCYs (2H, m), 6.946 (2H, t, J = 8.824 Hz), 6.731 (1H, dd, J1 = 7.721 Hz,
and As and with a commercial available precursor susceptible to J2 = 1.471Hz), 6.605 (1H, t, J = 7.721), 6.426 (1H, dd, J1 = 1.471 Hz,
decomposition.
J2 = 8.089), 5.940 (1H, d, J = 7.354), 3.997 (3H, s), 3.018–3.127
(2H, m), 2.777–2.831 (2H, m), 2.588–2.642 (2H, m), 2.047 (3H, s),
1.955–2.102 (1H, m), 1.677–1.813 (2H, m), 1.423–1.616 (2H, m),
1.239–1.382 (2H, m), 1.102 (9H, s).
Materials and methods
Chemicals were purchased from ACROS Sport GmbHBenzstr
(Renningen, Germany), Fluka, Denmark Sigma-Aldrich (Denmark
A/S Copenhagen, Denmark), Tocris Bioscience (Moorend Farm
Avenue, Bristol, United Kingdom) or Merck KGaA (Frankfurter
Strabe, Darmstadt, Deutschland). Unless otherwise stated, all
chemicals were used without further purification. For solid phase
extraction, Sep-PakW-QMA-and Sep-PakW-C18-cartridges were
used. TLC was performed using plates from Merck (silicagel
60F254 and alumina oxide 60 F254). 1H NMR spectra were
recorded using a Bruker AC 300. Chemical shifts are quoted as
d-values (ppm) downfield from tetramethylsilane. Field desorption
(FD) mass spectra were recorded using a Finnigan MAT90 spec-
trometer. Analytical and preparative HPLC were performed on a
Dionex (1027 Old York Rd. Ringoes NJ 08551-1054 USA) system
consisting of a pump P680A pump, a UVD 170U detector and a
Scansys radiodetector. Syntheses were carried out in a commer-
cially available MW oven (CEM LabMate, Corp. 45 Post Irvine, CA,
USA). [18F]Fluoride was produced via the 18O(p,n)18F reaction by
bombardment of an isotopically enriched [18O]water target with
an 11-MeV proton beam in a CTI (Middlebrook, PikeKnoxville, TN,
USA) Eclipse cyclotron. All radioactive syntheses were carried out
in an automated Scansys radiochemistry module.
(2-Methoxy-3-hydroxyphenyl)-(1-(2-p-fluorophenylethyl)-piperidine-
4-yl)-methyl acetate (4)
A solution of (3) (1.4 g, 2.18 mmol) and ammonium fluoride
(300 mg, 8 mmol) in dry MeOH was heated under reflux for
30 min. After cooling to room temperature, the solution was eva-
porated in vacuo, dissolved in H2O (30 mL), extracted with
dichloromethane (DCM) (3 Â 50 mL), washed with brine, dried
(Na2SO4) and evaporated. CC (DCM/MeOH 8:1) afforded (4) as
1
colourless crystals (667 mg, 1.66 mmol, 76%). H NMR (300 MHz,
CDCl3) d [ppm] = 7.202–7.250 (2H, m), 7.069 (2H, t, J = 8.824 Hz),
6.883 (1H, t, J = 8.089), 6.782 (1H, d, J = 8.089), 6.660 (1H, d,
J = 7.721), 5.762 (1H, d, J = 7.721), 3.749 (3H, s), 2.828–3.049 (2H,
m), 2.639–2.753 (2H, m), 2.406–2.566 (2H, m), 2.017 (3H, s), 1.883–
2.014 (1H, m), 1.609–1.755 (2H, m), 1.114–1.358 (4H, m).
(2-Methoxy-3-(tosyloxyethyloxy)phenyl)-(1-(2-p-fluorophenylethyl)-
piperidine-4-yl)-methyl acetate (5)
A solution of (4) (400 mg, 1 mmol), Cs2CO3 (390 mg, 1.2 mmol)
and ethyleneditosylate (1.85 g, 5 mmol) in dry MeCN (30 mL)
was heated under reflux for 16 h. The solvent was removed in
vacuo; the residue dissolved in H2O and extracted with Et2O.
The combined organic phases were washed with brine, dried
(Na2SO4) and evaporated. CC (DCM/ MeOH 6:1) afforded (5) as
colourless oil (276 mg, 0.46 mmol, 46%). 1H NMR (300 MHz,
CDCl3) d [ppm] = 7.780–7.807 (2H, d, J = 8.089 Hz), 7.314–7.343
(2H, d, J = 8.457), 7.126 (2H, dd, J1 = 8.457, J2 = 5.515), 6.839–6.991
(4H, m), 6.724 (1H, dd, J1 = 8.089, J2 = 1.471), 5.902 (1H, d,
J = 6.986), 4.353–4.385 (2H, m), 4.159–4.188 (2H, m), 3.789
(3H, s), 2.962–3.162 (2H, m), 2.765–2.871 (2H, m), 2.544–2.674
(2H, m), 2.429 (3H, s), 1.981–2.102 (2H, m), 2.044 (3H, s),
1.699–1.872 (2H, m), 1.341–1.652 (3H, m) MS (FD) m/z (% rel.Int.):
598.7 (100.0 [M]+), 599.7 (34.2 [M+ 1]+).
Chemistry
(3-((Tert-butyldiphenylsilyl)oxy)-2-methoxyphenyl)(piperidin-4-yl)
methanol (1) was synthesised as previously described.10
(2-Methoxy-3-(t-butyldiphenylsilyloxy)-phenyl)-(1-(2-p-fluorophe-
nylethyl)-piperidine-4-yl)-methanol (2)
The free amine (1) (1.5 g, 3.15 mmol), p-fluorophenethylbromide
(0.64 g, 3.15 mmol), NaI (0.7 g, 4.65 mmol) and NaHCO3 (0.53 g,
6.3 mmol) were dissolved in dry DMF (25 mL) and stirred at
85 ꢀC for 2 h. After evaporation of the solvent, the residue
was taken up in NH4OH (30 mL) and extracted with EtOAc
(3 Â 40 mL). The combined organic extracts were washed with
brine, dried (Na2SO4) and evaporated. Column chromatography
(CC) (chloroform/methanol 8:1) yielded (2) (1.32 g, 2.21 mmol,
70%). 1H NMR (300 MHz, CDCl3) d [ppm] = 7.732–7.681 (4H,
m), 7.416–7.301 (6H, m), 7.156–7.110 (2H, m), 6.969–6.911
(2H, t), 6.787–6.762 (1H, d), 6.642–6.589 (1H, d), 6.418–6.391
1H, d), 4.622–4.595 (1H, d), 3.968 (3H, s), 3.113–3.077 (1H, d), 2.975–
2.932 (1H, d), 2.806–2.752 (2H, m), 2.571–2.517 (2H, m), 2.101–1.924
(3H, m), 1.650–1.639 (1H, m), 1.554–1.215 (3H, m), 1.102 (9H, s) MS
(FD) m/z (% rel Int.): 597.6 (100.0 [M]+); 598.6 (77.39 [M + 1]+); 599.6
(24.91 [M + 2]+); 595.6.
(3-(2-Fluoroethoxy)-2-methoxyphenyl)(1-(4-fluorophenethyl)piperi-
din-4-yl)methanol (MH.MZ, (1))
MH.MZ was synthesised as reported by Herth et al.10
Production of [18F]MH.MZ
To an aqueous [18F]fluoride solution (1400–1600 MBq), KryptofixW222
(ABX GmbH, Heinrich-Glaeser-Strasse, Radeberg, Germany)
(15 mg), 15-mL potassium carbonate (1 N) and 1-mL acetonitrile
were added. The mixture was dried in a stream of nitrogen at
80 ꢀC. The drying procedure was repeated three times until the
reaction mixture was absolutely dry. The dried KryptofixW222/
[
18F]fluoride complex was then dissolved in 1-mL acetonitrile,
(2-Methoxy-3-(t-butyldiphenylsilyloxy)-phenyl)-(1-(2-p-fluoropheny-
lethyl)-piperidine-4-yl)-methyl acetate (3)
and 3-mg (1-(4-fluorophenethyl)piperidin-4-yl)(2-methoxy-3-
(2-(tosyloxy)ethoxy)phenyl)methyl acetate (5, 0.005 mmol))
To a solution of (2) (1.32 g, 2.2 mmol) in dry THF (25 mL), acetic was added. The resulting solution was heated at 80 ꢀC under
acid anhydride (420 mL, 4.4 mmol) was added dropwise at room stirring in a sealed vial for 25 min. 0.1 mL of a 0.05 M NaOMe
temperature. After stirring for 18 h at room temperature, the solution was added, and the solution was subsequently
www.jlcr.org
Copyright © 2012 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2012

M. M. Herth et al.
cooled to room temperature (RT) by nitrogen cooling. The 18F-nucleophilic labelling procedure, and accordingly, automation
solution was stirred for further 3 min and then quenched is easier.
with 3.5 mL of HPLC eluent. Purification of the crude prod-
uct was accomplished using HPLC (Luna 5 m C18(2) 100 Å,
Acknowledgements
250 Â 10.00 mm, 5 mm; acetonitrile/buffer (0.25M NH4Ac ad-
justed to pH= 5 with acetic acid) 35:65, flow rate: 6 mL/min,
The authors wish to thank the staff at the PET and Cyclotron
RT: 510 s ([18F]MH.MZ). After diluting the collected HPLC frac-
unit for expert technical assistance. Financial support by Intra
European Fellowship (MC-IEF-275329), the Faculty of Health at
University of Copenhagen and the Lundbeck Foundation are
gratefully acknowledged. The authors also wish to thank Hanne
D. Hansen and Anders Ettrup for fruitful discussions.
tion containing [18F]MH.MZ with water, the product was
trapped on an activated Sep-PakW-C18-cartrige and subse-
quently eluted with 3 mL of ethanol into 12 mL of phosphate-
buffered saline.
Determination of radiochemical purity and specific activity
Conflict of Interest
The radiotracer preparation was visually inspected for clarity,
absence of colour and particles. Chemical and radiochemical
purities were also assessed on this aliquot by TLC and HPLC
analysis.
The authors did not report any conflict of interest.
References
[18F]MH.MZ: Luna 150 Â 4.6 mm 5 C18(2) ((0.25 M NH4Ac
adjusted to pH = 5 with acetic acid) 35:65, flow rate: 1 mL/min,
RT: [18F]F- = 2.4 min; [18F]MH.MZ = 6.22 min; the intermediate
[18F](5) = 15.33 min; precursor (5) = 11.98 min; SiO2-TLC: eluent:
CHCl3/MeOH, 8:1, Rf: [18F]MH.MZ: 0.36, Rf: intermediate [18F](5):
0.85 and Rf: [18F]fluoride ion: 0.0).
[1] M. Naughton, J. B. Mulrooney, B. E. Leonard, Hum. Psychopharmacol.
Clin. Exp. 2000, 15, 397–415.
[2] A. Abbas, B. L. Roth, Expert Opin. Pharmacother. 2008, 9, 3251–3259.
[3] J. Gonzalez-Maeso, N. V. Weisstaub, M. Zhou, P. Chan, L. Ivic, R. Ang,
A. Lira, M. Bradley-Moore, Y. Ge, Q. Zhou, S. C. Sealfon, J. A. Gingrich,
Neuron 2007, 53, 439–452.
[4] L. M. Paterson, B. R. Kornum, D. J. Nutt, V. W. Pike, G. M. Knudsen,
Med. Res. Rev. 2011, DOI: 10.1002/med.20245.
[5] H. Watabe, M. A. Channing, M. G. Der, H. R. Adams, E. Jagoda,
P. Herscovitch, W. C. Eckelman, R. E. Carson, J. Cereb. Blood Flow
Metab. 2000, 20, 899–909.
Specific activity (As) of the radiotracers were calculated from
three consecutive HPLC analyses (average), determined by the
area of the UV absorbance peak corresponding to the radiola-
belled product on the HPLC chromatogram and compared with
a standard curve relating mass to UV absorbance.
[6] P. Z. Tan, R. M. Baldwin, T. Fu, D. S. Charney, R. B. Innism, J. Labelled
Compd. Rad. 1999, 42, 457–467.
[7] U. Mühlhausen, J. Ermert, M. M. Herth, H. H. Coenen. J. Labelled
Compd. Rad. 2009, 52, 6–12.
Production of [18F]MH.MZ via microwave heating
[8] M. M. Herth, F. Debus, M. Piel, M. Palner, G. M. Knudsen, H. Luddens,
F. Rösch, Bioorg. Med. Chem. Lett. 2008, 18, 1515–1519.
[9] F. Debus, M. M. Herth, M. Piel, H. G. Buchholz, N. Bausbacher,
V. Kramer, H. Lüddens, F. Rösch, Nucl. Med. Biol. 2010, 37, 487–495.
[10] M. M. Herth, V. Kramer, M. Piel, M. Palner, P. J. Riss, G. M. Knudsen, F.
Rösch, Bioorg. Med. Chem. 2009, 17, 2989–3002.
18F-Labelling kinetics of the intermediate ([18F](5)) of [18F]MH.MZ
in MeCN via MW heating: precursor concentration 3 mg/1 mL
MeCN; ramp time 120 s; Power max (300 W with cooling); hold
(heating until set temperature and hold set temperature).
[11] M. M. Herth, M. Piel, F. Debus, U. Schmitt, H. Luddens, F. Rösch, Nucl.
Med. Biol. 2009, 36, 447–454.
[12] C. Lemaire, R. Cantineau, M. Guillaume, A. Plenevaux, L. Christiaens, J.
Nucl. Med. 1991, 32, 2266–2272.
[13] A. de la Hoz, A. Diaz-Ortiz, A. Moreno, Chem. Soc. Rev. 2005, 34,
164–178.
[14] S. Stone-Elander, N. Elander, J. Labelled Compd. Rad. 2002, 45,
Conclusion
We developed a method for direct radiofluorination of the 5-HT2A
receptor antagonist [18F]MH.MZ and show that the direct does not
have any significant benefits over the indirect labelling method.
However, only one HPLC separation is necessary for the direct
715–746.
J. Label Compd. Radiopharm 2012
Copyright © 2012 John Wiley & Sons, Ltd.
www.jlcr.org