Radiosynthesis of the HIV integrase inhibitor [18F]MK-0518 (Isentress)

Article abstract of DOI:10.1002/jlcr.1778

The human immunodeficiency virus integrase inhibitor, [¹⁸F]MK- 0518, was prepared via a three-step, one-pot radiosynthesis. [ ¹⁸F]4-Fluorobenzylamine was produced from the fluorination of 4-cyano-N,N,N-trimethylammonium triflate with [¹⁸F]fluoride and reduction with borane methylsulfide complex in 50-68% radiochemical yield. The final step, the coupling of [¹⁸F]4-fluorobenzylamine with an ester coupling partner, achieved an overall uncorrected radiochemical yield after HPLC purification of ～2%, based on the starting [¹⁸F]fluoride. In a typical run, the total synthesis time was about 90 min and gave 0.37-1.74 GBq (10-47 mCi) of [¹⁸F]MK-0518. The radiochemical purity of [ ¹⁸F]MK-0518 was >98% and the specific activity was 243-1275 Ci/mmol (EOS, n = 4). A convenient three-step, one-pot radiosynthesis of [ ¹⁸F]MK-0518 via [¹⁸F]4-fluorobenzylamine has been developed, giving sufficient quantities of [¹⁸F]MK-0518 for animal positron emission tomography studies. Copyright

Full text of DOI:10.1002/jlcr.1778

Research Article
Received 14 December 2009,
Revised 12 March 2010,
Accepted 17 March 2010
Published online 11 May 2010 in Wiley Interscience
(www.interscience.wiley.com) DOI: 10.1002/jlcr.1778
Radiosynthesis of the HIV integrase inhibitor
[18F]MK-0518 (Isentress)
Ã
Wenping Li,a Wayne Thompson,b Thorsten Fisher,b John S. Wai,b
Daria Hazuda,c H. Donald Burns,a and Terence G. Hamilla
The human immunodeficiency virus integrase inhibitor,
[
18F]MK-0518, was prepared via a three-step, one-pot
radiosynthesis. [18F]4-Fluorobenzylamine was produced from the fluorination of 4-cyano-N,N,N-trimethylammonium
triflate with [18F]fluoride and reduction with borane methylsulfide complex in 50–68% radiochemical yield. The final step,
the coupling of [18F]4-fluorobenzylamine with an ester coupling partner, achieved an overall uncorrected radiochemical
yield after HPLC purification of ꢀ2%, based on the starting [18F]fluoride. In a typical run, the total synthesis time was about
90 min and gave 0.37–1.74 GBq (10–47 mCi) of [18F]MK-0518. The radiochemical purity of [18F]MK-0518 was498% and the
specific activity was 243–1275 Ci/mmol (EOS, n = 4). A convenient three-step, one-pot radiosynthesis of [18F]MK-0518 via
[
tomography studies.
18F]4-fluorobenzylamine has been developed, giving sufficient quantities of [18F]MK-0518 for animal positron emission
Keywords: HIV integrase inhibitor; PET imaging; Isentress; fluorine-18
O
N
Introduction
OH
O
F
N
The human immunodeficiency virus (HIV) is a retrovirus that
infects cells of the human immune system, destroying or
impairing their function. The most advanced stage of HIV
infection is acquired immunodeficiency syndrome (AIDS). It is
estimated that over 33 million people are living with HIV/AIDS
worldwide, of which 1.2 million were in the United States at the
end of 2008.1 Although decades of research and the successful
development of combination antiretroviral therapies have
transferred HIV infection from a fatal to a chronic life-
threatening disease, HIV remains one of the most serious health
problems in the world.2–4
HIV encodes three enzymes: reverse transcriptase (RT),
protease (PR), and integrase (IN), each of which play essential
roles in viral replication.5,6 The standard HIV/AIDS treatment
models based on nucleosidic and non-nucleosidic RT inhibitors
and/or PR inhibitors have been limited due to the multidrug-
resistant HIV infection and cross-resistance to agents within a
class.7 Therefore, novel drugs targeting other steps in the HIV-1
life cycle are needed.
HIV-1 IN catalyzes the insertion of the proviral DNA into the
host genome (strand transfer) and integration is required for both
the stable maintenance of the viral genome and for efficient viral
gene expression and replication.8–10 IN inhibitors are an attractive
new class of HIV drugs. Isentress (MK-0518, Raltegravir), shown in
Figure 1, is the first IN inhibitor approved by the Food and Drug
Administration for use in the treatment of HIV-1 infection in
combination with other antiretroviral agents.11–14
N
N
H
H
N
N
O
O
Figure 1. Chemical structure of MK-0518 (Isentress, Raltegravi) (Reference:12).
amine, a very useful intermediate in F-18 chemistry, which can by
synthesized by a variety of routes.15–20 Generally, it is prepared
by the reduction of [18F]4-fluorobenzonitrile. Lithium aluminum
hydride (LiAlH4) has been reported as the preferred reducing
agent because of its rapid reaction and quantitative yield.16
However, the use of LiAlH4 is less convenient due to the required
anhydrous reaction conditions, the formation of large amounts
of aluminum salts, and the troublesome purification. Borane
reagents are very effective for the conversion of nitriles to amines
and have advantages over LiAlH4.21 For the work described in
this study, borane methylsulfide complex (BH3-Me2S, BMS) was
selected as the reducing agent, the procedures described in the
literature were modified and a one-pot synthesis of [18F]MK-0518
via [18F]4-fluorobenzylamine was developed.16,18
aImaging Research, Merck Research Laboratories, West Point, PA, USA
bMedicinal Chemistry, Merck Research Laboratories, West Point, PA, USA
cInfectious Diseases, Merck Research Laboratories, West Point, PA, USA
The goal of this work was to synthesize [18F]MK-0518 to enable
the non-invasive study of MK-0518 biodistribution in rhesus
monkeys using PET. The planned synthesis of [18F]MK-0518,
shown in Scheme 1, would involve using [18F]4-fluorobenzyl
*Correspondence to: Wenping Li, Imaging Research, Merck Research Labora-
tories, WP 44D-2, West Point, PA 19486, USA.
E-mail: wenping_li@merck.com
J. Label Compd. Radiopharm 2010, 53 517–520
Copyright r 2010 John Wiley & Sons, Ltd.

W. Li et al.
higher in a completely cooled reaction mixture (room tempera-
ture) compared with warmer solutions in trial reactions.
Reduction with LiAlH4 needs to be performed in harsh
conditions, such as refluxing THF at 120–1401C for 2 min,
followed by quenching the excess of LiAlH4 and Sep-pak
purification.15–17,19 This procedure is somewhat troublesome
and time consuming. The use of borane reagents, such as
borane-THF and borane-Me2S (BMS), have been described to be
very effective for the conversion of [18F]fluorobenzonitrile to
[18F]2 under mild conditions.18 Borane-Me2S in THF was selected
over LiAlH4 as the reducing agent not only to avoid the
complicated purification of excess reducing agent but also to
prevent the formation of by-products.
Initially, the purification of [18F]2 with a C18 Light Sep-Pak was
attempted. The reduction mixture was quenched with 5 M HCl
and basified with 5 M NaOH and water. The basic solution was
loaded onto two prewashed C18 Light Sep-Pak cartridges, and
the cartridges were washed with 0.01 N NaOH. [18F]2 was finally
eluted from the cartridges in ethanol through a drying column
containing a mixture of potassium carbonate and sodium sulfate
(60:40). Using this procedure, pure [18F]2 was obtained as a dry
ethanolic solution with an uncorrected radiochemical yield of
ꢀ25% based on the initial activity of [18F]FÀ.
NH2
CN
-
1) [18F]KF/K222/DMSO
2) BH3-Me2S/THF
CF3SO3
18F
N+
[
18F]2
1
N
N
H
N
O
45W, 80 oC,
+
O
O
N
N
10 min
MeO
O
OH
3
N
N
H
N
O
18F
O
O
N
N
H
N
O
OH
[
18F]MK-0518
Scheme 1. Radiosynthesis of [18F]MK-0518.
A significant amount of radioactivity (ꢀ30%) remained on the
Sep-Pak cartridges and drying column. Increasing the elution
volume of ethanol resulted in more radioactivity eluting from
the cartridge, suggesting the radioactivity retained on the Sep-
Results and discussions
Radiochemistry
The three-step, one-pot radiosynthesis of [18F]MK-0518 is Pak cartridge was due to [18F]2. This purification sequence
outlined in Scheme 1. In the first step, [18F]4-fluorobenzonitrile required about 25 min but resulted in large volumes of elution
was synthesized via
a nucleophilic aromatic substitution solvent making it difficult to perform an automated synthesis of
reaction. 4-Cyano-N,N,N-trimethylammonium triflate was chosen [18F]2. To shorten the synthesis time and improve the radio-
as the starting material, due to trimethylammonium being the chemical yield, omitting the Sep-Pak purification step was
preferred leaving group over halo or nitro substituents.16–20 In investigated. This led to decreasing the overall synthesis time by
previous reports, the reaction of 1 with [18F]KF-K222 complex was nearly 15 min and increasing the available activity of [18F]2,
performed by heating 1 in dimethyl sulfoxide (DMSO) for 10 to ultimately improving the radiochemical yield of [18F]MK-0518.
20min at temperatures ranging from 120 to 1801C with a
The final step, coupling [18F]2 with ester 3 was performed in
radiochemical yields of 60–85%.16,18–20 Kuhnast et al. reported the the same reaction vial. Previous model reactions using
preparation of [18F]4-fluorobenzonitrile by microwave heating at unlabeled materials had been carried out to determine the best
100 W for 1 min resulting in a similar radiochemical yield.17 conditions for the final coupling step. The coupling of methyl
However, no temperature was reported for the substitution.
ester 3 with unlabeled 4-fluorobenzylamine was explored under
To determine the optimum reaction conditions in our hands, different reaction conditions including solvents (ethanol and
this reaction was carried out under various temperatures DMSO), molar ratios of ester:amine (1:1, 1:0.5, 1:0.1), reaction
(110–1601C), input power (75, 100 W), and reaction times times (30, 60 min), and temperatures (80, 1001C), based on the
(1–5 min). All reactions were carried out in DMSO. Similar synthetic route reported by Summa et al.12 The coupling of 3
radiochemical yields (75–85%) were observed when heated at was achieved with a chemical yield of 80% or greater within
1201C or higher temperatures. Lower temperatures were 30 min under all conditions tested, and no differences in the
preferred to prevent the loss of volatile [18F]4-fluorobenzonitrile. chemical yield were observed between different solvents or
Heating for 3 min provided the highest radiochemical yields temperatures. Therefore, the final coupling reaction was carried
(X85%) when varying the reaction time. Heating at 75 W or out at 801C in either ethanol or DMSO for 30 min. However,
100 W was also tested, and no difference in the radiochemical under no-carrier added conditions, the coupling of [18F]2 in
yield was observed. Therefore, the substitution was carried out DMSO provided a slightly higher yield than in ethanol with a
in DMSO using microwave heating at 75 W, 1201C for 180 s, shorter reaction time 10 min. Longer reaction times (30 min) did
providing the desired [18F]4-fluorobenzonitrile. Typically, the not improve the yield of [18F]MK-0518. The coupling [18F]2 with
fluorination reaction under these conditions resulted in a ester 3 at 45 W, 801C for 10 min in DMSO provided the highest
radiochemical yield of X80% from [18F]fluoride, which was radiochemical yield among the conditions used, providing
comparable with the methods reported in the literature.16–20
[18F]MK-0518 in an overall uncorrected radiochemical yield
The second step in this sequence was the conversion of (decay uncorrected) of 1–4%. The trace from the final
[18F]4-fluorobenzonitrile to [18F]2 with BMS. After cooling the preparative HPLC purification of [18F]MK-0518 is shown in
crude reaction from step 1 to room temperature, the crude Figure 2. The desired product eluted at 15.5 min (Panel a) and
[18F]4-fluorobenzonitrile was treated with an excess of BMS. only the fractions across the narrow peak were collected to
[18F]2 was formed with a 50–68% radiochemical purity based on avoid the contamination of a UV by-product, eluting at 16.5 min
analytical HPLC. It was noticed that this conversion rate was (Panel b). The final [18F]MK-0518 displayed no by-product UV
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 517–520

W. Li et al.
Figure 2. HPLC chromatograms (radioactivity and UV absorbance vs time) of the
purification of [18F]MK-0518 with a semi-preparative reversed-phase HPLC column:
(a) radioactivity and (b) UV absorbance at 254 nm.
Figure 4. Chromatograms of analytical HPLC of [18F]MK-0518 co-injected with
authentic standard: (a) radioactivity and (b) UV absorbance at 332.6 nm. The small
shoulder peak at ꢀ17 min of RT was attributed from the authentic standard, and
was not observed in HPLC analysis of no-carrier added [18F]MK-0518.
Gilson (Worthington, OH) 233XL liquid handler and microwave
reactions were carried out in a Resonance Instruments Model
521A Microwave Cavity with integrated temperature control.
A Capintec CRC-35R dose calibrator was used for radioactivity
measurements. Waters C-18 Light Sep-pak cartridges were pre-
activated with ethanol (5 ml), followed by sequential washing
with water (10 ml) and 0.01 N NaOH (10 ml) before use.
The [18F]MK-0518 was purified on a semi-preparative reverse-
phase HPLC system equipped with a Waters (Milford, MA) 600
pump and 600 Controller using a Waters XBridge C18 column
(5 mm, 10 Â 150 mm). The mobile phase consisted of acetonitrile
(CH3CN) containing 0.1% TFA (A) and phosphate buffer (10 mM
Na2HPO4, pH 7.4) (B) and a gradient method was used. A linear
gradient of 10% A to 30%A over 20 min and 30% A to 60% A
over 5 min, holding at 60% A for 10 min with a run time of
35 min at 4 ml/min was used. The preparative run was
monitored at 254 nm with an Amersham Bioscience (Piscataway,
NJ) UV-M II detector and a Bioscan (Missisauga, Ont., Canada)
FlowCount radioactivity detector. [18F]MK-0518 eluted at 15 min
and its radiochemical purity and identity was determined with
analytical HPLC by coinjection with the authentic standard.
Analytical reverse-phase HPLC was accomplished on a Waters
(Milford, MA) 600E chromatography system with a Waters 991
photodiode array detector and a Bioscan (Missisauga, Ontario,
Canada) FlowCount radioactivity detector. All samples were
analyzed on a Waters X-Terra C18 column (5 mm, 4.6 Â 150 mm).
The mobile phase was CH3CN (0.1% TFA) (A) and phosphate
buffer (10 mM Na2HPO4, pH 7.4) (B). Gradient elution consisted
of a linear gradient in 20 min from 10/90 to 30/70 and another
linear gradient followed in 5 min from 30/70 to 60/40 at
1 ml/min. A 5-min washing-out at 60/40 was followed to the end
of run (30 min run time). Both [18F]4-fluorobenzonitrile
(Rt = 27 min) and [18F]2 (Rt = 22 min) were confirmed with
commercial standards with UV absorbance at 254 nm, respec-
tively. [18F]MK-0518 co-eluted with the authentic standard at
15.3 min at 332.6 nm of UV maximum absorption wavelength.
Specific activities for [18F]MK-0518 were determined by using
analytical HPLC using the same system described above against
a calibration curve prepared with the cold standard.
Figure 3. Chromatograms of analytical HPLC of no-carrier added [18F]MK-0518:
(a) radioactivity and (b) UV absorbance at 332.6 nm.
peak detected on the analytical HPLC chromatogram (Figure 3).
The total synthesis time was about 90 min and gave
0.37–1.74 GBq (10–47 mCi) of pure [18F]MK-0518 with a specific
activity of 5877480 Ci/mmol (n = 4) and
a radiochemical
purity498%. The identity of [18F]MK-0518 was confirmed by
co-elution with the authentic standard using analytical HPLC
analysis (Figure 4).
Experimental
Materials and methods
General
BMS (2 M in THF), 4-nitrobenzonitrile, 4-fluorbenzylamine, and all
solvents were purchased from Aldrich Chemical Co. and were
used without further purification. Compounds 1, 3 and MK-0518
were provided by Merck Research Laboratories. [18F]FÀ was
obtained from Siemens Biomarker Solutions (North Wales, PA)
on an anion exchange resin transported to the radiochemistry
laboratory. Radiochemical procedures were performed using a
J. Label Compd. Radiopharm 2010, 53 517–520
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

W. Li et al.
Radiochemistry
10–47 mCi (0.37–1.74 GBq) of [18F]MK-0518 with a specific
activity ranging from 243 to 1275 Ci/mmol (n = 4) and radio-
chemical purity498%. The results of the animal PET studies will
be the subject of another study and published elsewhere.
The [18F]fluoride was obtained on an anion exchange resin from
Siemens Biomarker Solutions, transported to the radiochemistry
laboratory, and was eluted with a mixture (0.5 ml) of 80% MeCN:
20% oxalate aqueous solution (a mixture of 0.05 ml of 200 mg
K2C2O4/3 mg K2CO3/5 ml H2O, 0.25 ml H2O, and 1.2 ml MeCN)
and azeotropically dried under an argon stream adapted from a
reported procedure.22 Briefly, the eluted [18F]FÀ was transferred
into a 1 ml v-vial containing Kryptofix 222 (0.2 ml, 36 mg/ml in
MeCN). The vial was vented with an 18G1 syringe needle and
the mixture was dried under an argon stream using microwave
heating at 40 W, 801C for 3 min. Additional aliquots of
acetonitrile (3 Â 0.5 ml) were added and heated to dryness.
4-[18F]fluorobenzylamine ([18F]2). The synthesis of [18F]4-
fluorobenzylamine ([18F]2) was carried out with some modifica-
tion as previously described.17,18
Fluorination: A precursor solution of 1 (5 mg) in anhydrous
DMSO (0.2 ml) was added to the dry cryptate of [18F]KF in a 1-ml
vial, the vent needle was removed, and the reaction mixture was
heated at 75 W, 1201C for 180 s in the microwave reactor system.
An aliquot (10 ml) was removed and analyzed by using analytical
HPLC as described above to determine the radiochemical yield.
Reduction: The reaction mixture was allowed to cool to room
temperature. BMS (400 ml, 2 M in THF) was added, and the
mixture was stirred first at room temperature for 5 min with an
argon flow. An aliquot (10 ml) was removed and analyzed to
determine the radiochemical purity by using the same analytical
HPLC method as described above. The THF was then removed
under argon flow with microwave heating (35 W, 651C for 5 min
and then at 40 W, 801C for 5 min).
Synthesis of [18F]MK-0518. Methyl ester 3 (ꢀ5 mg) in DMSO
(300 ml) was added to the crude reaction mixture. The mixture
was heated in the microwave (45 W, 801C) for 10 min. After
cooling for 1 min, the mixture was diluted with water (800 ml)
and loaded onto an Xbridge C-18 semi-preparative HPLC
column. The peak corresponding to [18F]MK-0518 eluting at
15 min was collected, and most of the solvent was evaporated,
and transferred into a sterile vial for animal studies. One typical
synthesis for animal PET study provided 47 mCi of [18F]MK-0518
with a specific activity of 1275 Ci/mmol at the end of synthesis
and a radiochemical purity498% in a radiochemical yield 4%.
References
[1] 2008 Report on the Global AIDS Epidemic, UNAIDS. Geneva,
Switzerland 2008.
[2] F. J. Palella, K. M. Delaney, A. C. Moorman, M. O. Loveless,
J. Fuhrer, G. A. Satten, D. J. Aschman, S. D. Holmberg, N. Engl. J.
Med. 1998, 338, 853–860.
[3] E. Wood, R. S. Hogg, B. Yip, P. R. Harrigan, M. V. O’Shaughnessy,
J. S. Montaner, AIDS 2003, 17, 711–720.
[4] G. Chene, J. A. Sterne, M. May, D. Costagliola, B. Ledergerber,
A. N. Phillips, F. Dabis, J. Lundgren, A. D’Arminio Monforte,
F. de Wolf, R. Hogg, P. Reiss, A. Justice, C. Leport, S. Staszewski,
J. Gill, G. Fatkenheuer, M. E. Egger; The Antiretroviral Therapy
Cohort Collaboration, Lancet 2003, 362, 679–686.
[5] E. Asante-Appiah, A. M. Skalka, Antiviral Res. 1997, 36, 139–156.
[6] P. hindmarsh, J. Leis, Microbiol. Mol. Biol. Rev. 1999, 63,
836–843.
[7] J. Cohen, Science 2002, 296, 2320–2324.
[8] Y. Pommier, A. A. Johnson, C. Marchand, Nat Rev. Drug Discov.
2005, 4, 236–248.
[9] D. J. Hazuda, P. Felock, M. Witmer, A. Wolfe, K. Stillmock,
J. A. Grobler, A. Espeseth, L. Gabryelski, W. Schleif, C. Blau,
M. D. Miller, Science 2002, 287, 646–650.
[10] R. L. LaFemina, C. L. Schneider, H. L. Robbins, P. L. Callahan,
K. Legrow, E. Roth, W. A. Schleif, E. A. Emini, J. Virol. 1992, 66,
7414–7419.
[11] M. Rowley, Progr. Med. Chem. 2008, 46, 1–28.
[12] V. Summa, A. Petrocchi, F. Bonelli, B. Crescenzi, M. Donghi,
M. Ferrara, F. Fiore, C. Gardelli, O. G. Paz, D. J. Hazuda, P. Jones,
O. Kinzel, R. Laufer, E. Monteagudo, E. Muraglia, E. Nizi, F. Orvieto,
P. Pace, G. Pescatore, R. Scarpelli, K. Stillmock, M. V. Witmer,
M. Rowley, J. Med. Chem. 2008, 51, 5843–5855.
[13] Y. Wang, N. Serradell, J. Bolos, E. Rosa, Drugs Fut. 2007, 32,
118–122.
[14] T. H. Evering, M. Markowitz, Drugs Today 2007, 43,
865–877.
[15] P. K. Garg, S. Garg, M. R. Zalutsky, Bioconjugate Chem. 1991, 2,
44–49.
[16] F. Dolle, F. Hinnen, F. Vaufrey, B. Tavitian, C. Crouzel, J. Label.
Comp. Radiopharm. 1997, 39, 319–330.
[17] B. Kuhnast, F. Hinnen, R. Boisgard, B. Tayitian, F. Dolle,
M. R. Kilbourn, G. L. Watkins, S. A. Toorongian, J. Label Comp.
Radiopharm. 2003, 46, 1093–1103.
[18] T. Haradahira, Y. Hasegawa, K. Furuta, M. Suzuki, Y. Watanabe,
K. Suauki, Appl. Radiat. Isot. 1998, 49, 1551–1556.
[19] I. Koslowsky, S. Shahhosseini, J. Wilson, J. Mercer, J. Label Comp.
Radiopharm. 2008, 51, 352–356.
Conclusion
[18F]MK-0518 has been successfully synthesized via the three-
step one-pot procedure described in this study, giving sufficient
quantities of [18F]MK-0518 for in vivo PET studies. The three-step,
one-pot radiosynthesis included nucleophilic fluorination, nitrile
reduction, and coupling in the same reaction vial and is simple
and convenient to use as no purifications of intermediates are
required. The total synthesis time was about 90 min and gave
[20] M. S. Haka, M. R. Kilbourn, G. L. Watkins, S. A. Toorongian, J. Label
Comp. Radiopharm. 1989, 27, 823–833.
[21] H. C. Brown, Y. M. Choi, S. Narasimhan, Synthesis 1981, 605–606.
[22] T. G. Hamill, S. Krause, R. Christine, C. Bonnefous, S. Govek,
T. J. Seiders, N. D. P. Cosford, J. Roppe, T. Kamenecka, S. Patel,
R. E. Gibson, S. Sanabria, K. Riffel, W. Eng, C. King, X. Yang,
M. D. Green, S. S. O’Malley, R. Hargreaves, H. D. Burns, Synapse
2005, 205–216.
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 517–520