Synthesis of [18F]Xeloda as a novel potential PET radiotracer for imaging enzymes in cancers

Nuclear Medicine and Biology 31 (2004) 1033–1041

www.elsevier.com/locate/nucmedbio

Synthesis of [18F]Xeloda as a novel potential PET radiotracer for imaging enzymes in cancers Xiangshu Feia, Ji-Quan Wanga, Kathy D. Millerb, George W. Sledgeb, Gary D. Hutchinsa, Qi-Huang Zhenga,* a

Department of Radiology, Indiana University School of Medicine, Indianapolis, IN 46202, USA Department of Medicine, Indiana University School of Medicine, Indianapolis, IN 46202, USA

b

Received 7 October 2003; received in revised form 16 December 2003

Abstract Xeloda (Capecitabine), a prodrug of antitumor agent 5-fluorouracil, is the first and only oral fluoropyrimidine to be approved for use as second-line therapy in metastatic breast cancer, colorectal cancer, and other solid malignancies. Fluorine-18 labeled Xeloda may serve as a novel radiotracer for positron emission tomography (PET) to image enzymes such as thymidine phosphorylase and uridine phosphorylase in cancers. The precursor 2⬘,3⬘-di-O-acetyl-5⬘-deoxy-5-nitro-N4-(pentyloxycarbonyl)cytidine (11) was synthesized from D-ribose and cytosine in 8 steps with approximately 18% overall chemical yield. The reference standard 5⬘-deoxy-5-fluoro-N4-(pentyloxycarbonyl)cytidine (Xeloda; 1) was synthesized from D-ribose and 5-fluorocytosine in eight steps with approximately 28% overall chemical yield. The target radiotracer 5⬘-deoxy-5-[18F]fluoro-N4-(pentyloxycarbonyl)cytidine ([18F]Xeloda; [18F]1) was prepared by nucleophilic substitution of the nitro-precursor with K18F/Kryptofix 2.2.2 followed by a quick deprotection reaction and purification with the HPLC method in 20 –30% radiochemical yields. © 2004 Elsevier Inc. All rights reserved. Keywords: [18F]Xeloda; Enzyme; Cancer; PET; Radiotracer; Fluorine-18

1. Introduction Xeloda (Capecitabine; Hoffman LaRoche, Nutley, NJ, USA), a prodrug of an earlier antitumor agent 5-fluorouracil (5-FU) prodrug, 5⬘-deoxy-5-fluorouridine (5⬘-DFUR; Furtulon, Nippon Roche, Tokyo, Japan), is the first and only oral fluoropyrimidine to be approved for use as second-line therapy in metastatic breast cancer, colorectal cancer, and other solid malignancies [1– 8]. 5-FU is one of the antitumor agents most frequently used for treating solid tumors, and is rapidly metabolized by the enzyme dihydropyrimidine dehydrogenase (DPD) in tumors. However, 5-FU is poorly tumor selective, and its therapy causes high incidences of toxicity in the bone marrow, gastrointestinal tract, central nervous system and skin. These have limited its usefulness. Xeloda, which is a relatively new antitumor agent with better efficacy and safety profiles, is not a DPD substrate, and is converted to 5-FU from the intermediates 5⬘-deoxy5-fluorocytidine (5⬘-DFCR) and 5⬘-DFUR intratumorally * Corresponding author. Tel: 317-278-4671; fax: 317-278-9711. E-mail address: [email protected] (Q.-H. Zheng). 0969-8051/04/$ – see front matter © 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.nucmedbio.2004.02.006

through its sequential metabolism by three enzymes highly expressed in the liver and tumors: carboxylesterase, cytidine deaminase and thymidine phosphorylase (dThdPase) [9,10], and preferentially locates in various types of cancer tissues. The susceptibility of human cancer xenografts to Xeloda correlated with the ratio of two enzymes in tumors, dThdPase and DPD, which respectively generate 5-FU from the intermediate 5⬘-DFUR and catabolize 5-FU to an inactive 5,6-dihydro-5-fluorouracil (FUH2). The efficacy of Xeloda should theoretically be optimized in tumors with high concentrations of dThdPase and low concentrations of DPD. The antitumor mechanism of Xeloda is shown in Fig. 1. The elevated levels of the enzyme dThdPase in tumors provide a potential target for positron emission tomography (PET) imaging of tumors. Xeloda is also likely a substrate for uridine phosphorylase, given that Xeloda is more akin to uridine and cytosine than it is to thymidine (a 2⬘-deoxynucleoside). Therefore, fluorine-18 labeled Xeloda may serve as an enzyme-based imaging agent [11] and enable noninvasive monitoring of the levels of tumor enzymes such as thymidine phosphorylase and uridine phosphorylase and tumor response to Xeloda therapy using PET imaging tech-

1034

Q.-H. Zheng et al. / Nuclear Medicine and Biology 31 (2004) 1033–1041

Fig. 1. Antitumor mechanism of Xeloda in humans.

niques. Recently, nucleoside tracers have been actively investigated as PET imaging agents for use in tumor detection and/or to monitor response to chemotherapy by a number of institutions, these previous attempts have been summarized in a recent mini-review [12], but the radiolabeling of Xeloda as a PET tracer is still not reported in the literature. In this paper, we describe the design and synthesis of [18F]Xeloda, for the first time.

2. Results and discussion 2.1. Chemistry The synthesis of intermediate 5-deoxy-1,2,3-tri-Oacetyl-D-ribofuranoside (7) as indicated in Scheme 1 was performed with the modifications according to procedures

reported in the literature [13]. The commercially available starting material D-ribose (2) was converted into methyl(2,3-O-isopropyliden-D-ribofuranoside) (3) through the reaction with acetone and methanol under acidic condition in a yield of 78%. Compound 3 was reacted with tosyl chloride in pyridine to give the tosylate methyl-(5-tosyl-5-deoxy2,3-O-isopropyliden-D-ribofuranoside) (4) in 84% yield. The tosylate 4 was converted to the iodide methyl-(5-iodo5-deoxy-2,3-O-isopropyliden-D-ribofuranoside) (5) through the reaction with sodium iodide in 97% yield. The reduction of iodide 5 by hydrogen gas afforded methyl-(5deoxy-2,3-O-isopropyliden-D-ribofuranoside) (6) in 95% yield, and the reaction was catalyzed by Pd/C. The hydrolysis of compound 6 by 1 N HCl aqueous solution followed by a protection reaction with acetic anhydride gave the intermediate 7 as a colorless oil in 99% yield. It was purified by recrystallization to give the pure product as a white solid.

Q.-H. Zheng et al. / Nuclear Medicine and Biology 31 (2004) 1033–1041

1035

Scheme 1. Synthesis of intermediate 5-deoxy-1,2,3-tri-O-acetyl-D-ribofuranoside.

The melting point (mp) of compound 7 measured has a deviation of 8°C from its previously reported value [13]. To explain the difference in mps, the HPLC method and the mixed mp method were performed. The isolated compound 7 was homogeneous by the HPLC method, and the upper mp of the two values was not depressed by the mixed mp method. In addition, compound 7 has 1H NMR data in

agreement with the indicated structure. Therefore, compound 7 could be that the crystals isolated were a special crystalline form with a different mp. The synthesis of the precursor 2⬘,3⬘-di-O-acetyl-5⬘-deoxy-5-nitro-N4-(pentyloxycarbonyl)cytidine (11) as outlined in Scheme 2 was performed with the modifications according to procedures reported in the literature [9]. Cy-

Scheme 2. Synthesis of precursor 2⬘,3⬘-di-O-acetyl-5⬘-deoxy-5-nitro-N4-(pentyloxycarbonyl)cytidine.

1036

Q.-H. Zheng et al. / Nuclear Medicine and Biology 31 (2004) 1033–1041

Scheme 3. Synthesis of reference standard 5⬘-deoxy-5-fluoro-N4-(pentyloxycarbonyl)cytidine (Xeloda).

tosine (8) was converted to 5-nitrocytosine (9) by HNO3/H 2SO4 in 64% yield. Compound 9 was reacted with 1,1,1,3,3,3-hexamethyldisilazane (HMDS) to provide 5-nitrocytosine trimethylsilyl derivative, which was directly reacted with the intermediate 7 with stannic tetrachloride in dichloromethane at 0°C to give glycosidation product 2⬘,3⬘di-O-acetyl-5⬘-deoxy-5-nitrocytidine (10) in 56% yield. The acylation of N4-amino group of compound 10 with n-pentyl chloroformate under basic condition afforded the precursor 11 in 84% yield. The synthesis of reference standard 5⬘-deoxy-5-fluoroN4-(pentyloxycarbonyl)cytidine (Xeloda; Capecitabine; 1) as shown in Scheme 3 was performed with the modifications according to procedures reported in the literature [9]. 5-Fluorocytosine (12) was reacted with 1,1,1,3,3,3-hexamethyldisilazane (HMDS) to provide 5-fluorocytosine trimethylsilyl derivative, which was directly reacted with the intermediate 7 with stannic tetrachloride in dichloromethane at 0°C to give glycosidation product 2⬘,3⬘-di-O-acetyl-5⬘deoxy-5-fluorocytidine (13) in 78% yield. Compound 13 was purified by the flash column chromatography. The mp of compound 13 recorded has a deviation of 110°C from its previously reported value in which compound 13 was purified by recrystallization [9]. As aforementioned for compound 7, the difference in mp’s can be explained by the HPLC method and the mixed mp method, and the indicated structure of compound 13 can be confirmed by its 1H NMR data. Likewise, compound 13 could be that the crystals isolated were a special crystalline form with a different mp. The acylation of N4-amino group of compound 13 with n-pentyl chloroformate under basic condition af-

forded 2⬘,3⬘-di-O-acetyl-5⬘-deoxy-5-fluoro-N4-(pentyloxycarbonyl)cytidine (14) in 73% yield. The alkaline hydrolysis of the acetyl group of compound 14 at 0°C gave the reference standard Xcloda 1 in 82% yield. The mp of compound 1 recorded has a deviation of 31°C from its previously reported value [9]. Likewise, compound 1 could be that the crystals isolated were a special crystalline form with a different mp. 2.2. Radiochemistry The synthesis of the target tracer 5⬘-deoxy-5-[18F]fluoroN -(pentyloxycarbonyl)cytidine ([18F]Xeloda; [18F]1) is outlined in Scheme 4. The nitro-precursor 11 was labeled by a conventional nucleophilic substitution [14 –16] with K 18 F/Kryptofix 2.2.2 in DMSO at 150°C for 15–20 min to provide a radiolabeling intermediate [18F]14. The radiolabeling reaction was monitored by analytical radio-HPLC method, in which we employed a analytical HPLC system [17–19] by using a Prodigy (Phenomenex) 5 ␮m C-18 column, 4.6 ⫻ 250 mm; 3:1:3 CH3CN: MeOH: 20 mM, pH 6.7 buffer (120 mL 100 mM KH2PO4 and 80 mL 100 mM K2HPO4 in 800 mL Zyza-tech H2O) mobile phase, 1.5 mL/min flow rate, and UV (240 nm) and ␥-ray (NaI) flow detectors. The radiolabeling mixture containing the intermediate [18F]14 was passed through a Silica Sep-Pak cartridge to remove Kryptofix 2.2.2 and nonreacted precursor 11 and K18F. The large polarity difference between [18F]14 and Kryptofix 2.2.2 and nonreacted precursor 11 and K18F permitted the use of a simple solid-phase extraction (SPE) technique [20 –22] for fast isolation of [18F]14 from the 4

Q.-H. Zheng et al. / Nuclear Medicine and Biology 31 (2004) 1033–1041

1037

Scheme 4. Synthesis of target tracer 5⬘-deoxy-5-[18F]fluoro-N4-(pentyloxycarbonyl)cytidine ([18F]Xeloda).

radiolabeling reaction mixture. The key part in this technique is a SiO2 Sep-Pak type cartridge, which contains ⬃0.5–2 g of adsorbent. The Sep-Pak was eluted with 15% MeOH/CH2Cl2 and the solvent was evaporated under high vacuum to give the compound [18F]14. The existence of the catalyst Kryptofix 2.2.2 and non-reacted precursor 11 and K18F would affect the deprotection reaction of [18F]14 and the quality control (QC) [23] of the labeled product [18F]1; therefore, they needed to be removed before [18F]14 was deprotected to give the target labeled product [18F]1. Moreover, the Silica Sep-Pak cartridge purification of [18F]14 from the radiolabeling reaction mixture before the deprotection reaction significantly increased the radiochemical yield of the target labeled product [18F]1 [24]. The acetyl protecting groups of [18F]14 were rapidly removed by treating with methanolic 6 N NaOH for 10 min followed by neutralization with 2 N HCl to provide [18F]1. The final radiolabeling mixture was purified by preparative HPLC method to provide pure target tracer [18F]1, in which we employed a preparative HPLC system [25] by using a Prodigy (Phenomenex) 5 ␮m C-18 column, 10 ⫻ 250 mm; 3:1:3 CH3CN: MeOH: 20 mM, pH 6.7 buffer (120 mL 100 mM KH2PO4 and 80 mL 100 mM K2HPO4 in 800 mL Zyza-tech

H2O) mobile phase, 5.0 mL/min flow rate, and UV (240 nm) and ␥-ray (NaI) flow detectors. The radiochemical yield of [18F]1 was 20 –30%, and the synthesis time including HPLC purification and formulation was 60 –70 min from end of bombardment (EOB). Compounds 11, 14, and 1 were used as nonradioactive reference compounds in both analytical and preparative HPLC methods to obtain retention time (Rt) parameters, which were used to monitor the radiolabeling and purification of [18F]1. Retention times in the analytical HPLC system were: Rt11 ⫽ 2.38 min, Rt14 ⫽ 4.02 min, Rt1 ⫽ 3.19 min. Retention times in the preparative HPLC system were: Rt11 ⫽ 3.77 min, Rt14 ⫽ 6.54 min, Rt1 ⫽ 4.58 min. The HPLC methods confirmed [18F]1 was actually isolated from the radiolabeling mixture, and it contained no other fluoro-nucleoside isomer and only trace amount of nonradioactive Xeloda that was produced by the reaction of the labeling precursor 11 with unlabeled fluoride. Chemical purity, radiochemical purity, and specific radioactivity were determined by the analytical HPLC method. The chemical purities of precursor 11, reference intermediate 14 and reference standard 1 were ⬎95%, the radiochemical purities of target radiotracer [18F]1 were ⬎99%, and the chemical purities of radiotracer [18F]1 were

1038

Q.-H. Zheng et al. / Nuclear Medicine and Biology 31 (2004) 1033–1041

⬃93%. The average (n ⫽ 3–5) specific radioactivities of radiotracer [18F]1 were 1.0 –1.2 Ci/␮mol at end of synthesis (EOS). To further verify the structure of the labeled Xeloda, the radiosynthesis with a small of moderate amount (15–20 mg) of carrier added was performed, and then the unlabeled product was isolated and characterized by proton-NMR spectrum. The results showed what we isolated as Xeloda has the same H NMR data as the authentic drug. This has excluded the possibility that the labeled compound we produced and isolated at the no-carrier-added level was not 6-[18F]Xcloda, an isomeric compound of 5-[18F]Xeloda ([18F]1). This compound could very conceivably be the product we isolated and it would be difficult to say that the HPLC methods, even with chemical correlations, could distinguish 5-[18F]Xeloda from this fluorinated regioisomer 6-[18F]Xeloda. Not even mass spectra would be useful in this situation. The carrier-added radiosynthesis design and discussion is based on the very possible prospect that nucleophilic substitution in this ring-system could be initiated by extended conjugate addition of fluoride at C-6 position and subsequent protonation at C-5 position, followed by deprotonation at C-6 position and elimination of nitrite at C-5 position. The carrier-added radiosynthesis results are consistent with the theoretical explanation and reaction mechanisms of nucleophilic substitution and conjugate addition that the nitro group at C-5 is a strong electron withdrawing group, and the electron density at C-6 is higher than at C-5, therefore, fluorine anion will prefer to attack at C-5 to occur nucleophilic substitution to form 5-[18F]Xeloda rather than at C-6 to occur conjugate addition to form 6-[18F]Xeloda, although the conjugate addition at C-6 is a potential competing reaction in comparison with the nucleophilic substitution at C-5. The radiosynthesis used a chiral precursor prepared from D-ribose. The removal of protecting group from the labeled precursor was done under mild basic conditions, which would not be expected to epimerize the anomeric center [26].

4. Experimental 4.1. General All commercial reagents and solvents were used without further purification unless otherwise specified. Melting points were determined on a MEL-TEMP II capillary tube apparatus and were uncorrected. 1H NMR spectra were recorded on a Bruker QE 400 NMR spectrometer using tetramethylsilane (TMS) as an internal standard. Chemical shift data for the proton resonances were reported in parts per million (␦) relative to internal standard TMS (␦ 0.0). The low resolution mass spectra (LRMS) were obtained using a Bruker Biflex III MALDI-TOF mass spectrometer, and the high resolution mass spectra (HRMS) measurements were obtained using a Kratos MS80 mass spectrometer, in the Department of Chemistry at Indiana University. Chromatographic solvent proportions are expressed on a volume: volume basis. Thin layer chromatography was run using Analtech silica gel GF uniplates (5 ⫻ 10 cm2). Plates were visualized by iodine or UV light. Normal phase flash chromatography was carried out on EM Science silica gel 60 (230 – 400 mesh) with a forced flow of the indicated solvent system in the proportions described below. All moisture-sensitive reactions were performed under a positive pressure of nitrogen maintained by a direct line from a nitrogen source. Analytical HPLC was performed using a Prodigy (Phenomenex) 5 ␮m C-18 column, 4.6 ⫻ 250 mm; 3:1:3 CH ⫺ 3CN: MeOH: 20 mM, pH 6.7 KHPO4 mobile phase, 1.5 mL/min flow rate, UV (240 nm) and ␥-ray (NaI) flow detectors. Semipreparative HPLC was performed using a Prodigy (Phenomenex) 5 ␮m C-18 column, 10 ⫻ 250 mm; 3:1:3 CH3CN: MeOH: 20 mM, pH 6.7 KHPO4⫺ mobile phase, 5.0 mL/min flow rate, UV (240 nm) and ␥-ray (NaI) flow detectors. Semi-prep C-18 SiO2 Sep-Pak type cartridge was obtained from Waters Corporate Headquarters, Milford, MA, USA. Sterile vented Millex-GS 0.22 ␮m filter unit was obtained from Millipore Corporation, Bedford, MA USA.

3. Conclusion 4.2. Synthesis of intermediate An efficient and convenient chemical and radiochemical synthesis of the intermediate, precursor, reference standard, and target tracer of Xeloda has been developed. Radiosynthesis produced [18F]Xeloda in amounts and purity suitable for further in vivo biological studies. Labeled product suitable for injection, with the specific radioactivities in a range of 1.0 –1.2 Ci/␮mol at EOS, can be obtained in 60 –70 min from EOB, including HPLC purification and formulation. These chemistry results provide the foundation for further evaluation of [18F]Xeloda as a new potential PET radiotracer for imaging enzymes in cancers.

4.2.1. Methyl-(2,3-O-isopropyliden-D-ribofuranoside) (3) A mixture of D-ribose (2, 19.25 g, 128.33 mmol) in acetone (75 mL), methanol (75 mL) and concentrated HCl (2 mL) was refluxed for 2 h. The reaction solution was diluted by H2O (200 mL) and extracted with CHCl3 (3 ⫻ 200 mL). The organic phase was dried over Na2SO4, evaporated and dried under vacuum to give a thick yellow oil 3 (20.50 g, 78%), Rf ⫽ 0.40 (EtOAc) (see ref. [13], boiling point (bp): 72–75°C/0.6 Torr), which was used for the next reaction without further purification.

Q.-H. Zheng et al. / Nuclear Medicine and Biology 31 (2004) 1033–1041

4.2.2. Methyl-(5-tosyl-5-deoxy-2,3-O-isopropyliden-Dribofuranoside) (4) Compound 3 (5.11 g, 25.05 mmol) was dissolved in pyridine (30 mL). Tosyl chloride (6.50 g, 34.09 mmol) was added to the solution. The mixture was stirred at room temperature for 18 h. The solvent was removed by evaporation under vacuum and the crude product was purified with flash column chromatography (1:4 EtOAc/hexane) to give a vacuum dried product as a white solid 4 (7.50 g, 84%), mp: 78 –79°C. 1H NMR (400 MHz, CDCl3): ␦ 1.29 (s, 3H, Me), 1.45 (s, 3H, Me), 2.46 (s, 3H, Me-Ar), 3.24 (s, 3H, OMe); 4.01– 4.03 (d, J ⫽ 7.2 Hz, 2H, CH2OTs), 4.29 – 4.33 (t, J ⫽ 6.8 Hz, 1H, H-C4), 4.53– 4.54 (d, J ⫽ 6.4 Hz, 1H), 4.59 – 4.61 (d, J ⫽ 6.4 Hz, 1H), 4.93 (s, 1H, H-Cl), 7.35–7.37 (d, J ⫽ 7.6 Hz, 2H, H-Ar), 7.80 –7.82 (d, J ⫽ 7.6 Hz, 2H, H-Ar). 4.2.3. Methyl-(5-iodo-5-deoxy-2,3-O-isopropyliden-Dribofuranoside) (5) A mixture of compound 4 (4.40 g, 12.30 mmol), Nal (5.03 g, 33.53 mmol) in 2-butanone (50 mL) was refluxed for 24 h. The solvent was evaporated and the residue was diluted in Et2O. Filtration gave a solution, which was evaporated and dried under vacuum to give a light yellow oil 5 (3.75 g, 97%), Rf ⫽ 0.60 (1:3 EtOAc/hexane). 1H NMR (400 MHz, CDCl3): ␦ 1.33 (s, 3H, Me), 1.49 (s, 3H, Me), 3.14 –3.19 (t, J ⫽ 10.4 Hz, 1H. CHI), 3.27–3.31 (dd, J ⫽ 6.0, 10.0 Hz, 1H, CHI), 3.38 (s, 3H, OMe), 4.43– 4.47 (dd, J ⫽ 6.0, 7.5 Hz, 1H, H-C4), 4.62– 4.64 (d, J ⫽ 5.6 Hz, 1H), 4.76 – 4.78 (d, J ⫽ 5.6 Hz, 1H), 5.06 (s, 1H, H-Cl). 4.2.4. Methyl-(5-deoxy-2,3-O-isopropyliden-Dribofuranoside) (6) Compound 5 (3.67 g, 11.69 mmol) was dissolved in methanol (100 mL) and Et3N (3.0 mL). 10% Pd/C (0.40 g) was added to the solution. The mixture was stirred at room temperature under H2 for 24 h. The solution was filtrated though Celite. The solvent was removed by evaporation and dried under vacuum to give a light yellow oil 6 (2.10 g, 95%), Rf ⫽ 0.50 (1:3 EtOAc/hexane) (see Ref. [13], bp: 73–75°C/12 Torr). 1H NMR (400 MHz, CDCl3): ␦ 1.27– 1.29 (d, J ⫽ 1.6, 6.8 Hz, 3H, Me-C5), 1.31 (s, 3H, Me), 1.48 (s, 3H, Me), 3.33 (s, 3H, OMe), 4.32– 4.37 (q, J ⫽ 6.8 Hz, 1H, H-C4), 4.50 – 4.52 (d, J ⫽ 6.0 Hz, 1H), 4.63– 4.65 (dd, J ⫽ 1.6, 6.0 Hz, 1H), 4.93– 4.94 (d, J ⫽ 1.6 Hz, 1H, H-Cl). 4.2.5. 5-Deoxy-1,2,3-tri-O-acetyl-D-ribofuranoside (7) A solution of the compound 6 (2.11 g, 11.22 mmol) in 1N HCl (1.5 mL) and H2O (35 mL) was refluxed for 2 h. The solvent was removed and the residue was dissolved in acetic anhydride (5.0 mL) and pyridine (50 mL). The reaction mixture was stirred at room temperature for 24 h. The solvent was removed and the residue was diluted by CH2Cl 2. The solution was washed with H2O. The organic phase was dried over Na2SO4 and the solvent was removed by evaporation to give a colorless oil 7 (2.89 g, 99%). It was

1039

purified by recrystallization in diisopropyl ether to give a vacuum dried product as a white solid 7, mp: 58 – 60°C (see Ref. [13], mp: 66 – 67°C). 1H NMR (400 MHz, CDCl3): ␦ 1.29 –1.31 (d, J ⫽ 6.4 Hz, 3H, Me), 2.00 –2.01 (d, J ⫽ 1.2 Hz, 3H, CH3CO), 2.02–2.03 (d, J ⫽ 1.2 Hz, 3H, CH3CO), 2.05 (s, 3H, CH3CO), 4.19 – 4.25 (q, J ⫽ 6.4 Hz, 1H, H-C4), 5.01–5.04 (dd, J ⫽ 1.2, 6.0 Hz, 1H), 5.26 –5.27 (d, J ⫽ 6.0 Hz, 1H), 6.05 (s, 1H, H-Cl). 4.3. Synthesis of precursor 4.3.1. 5-Nitrocytosine (9) The mixture of cytosine (8, 4.96 g, 44.68 mmol) in HNO 3 (30 mL) and concentrated H2SO4 (7.5 mL) was heated at 85°C for 1 day. The reaction mixture was cooled to room temperature and poured into ice water (300 mL). NaOH/H 2O solution was used to neutralize the water solution to pH 7. The solid was filtered and dried under vacuum to give a white solid 9 (4.50 g, 64%), mp: 290°C (decay). LRMS (EI, m/z): 156.98 (M⫹, 100%). HRMS (EI, m/z): calcd for C4H 4N4O3 156.0283, found 156.0282. 4.3.2. 2⬘,3⬘-Di-O-acetyl-5⬘-deoxy-5-nitrocytidine (10) Compound 9 (0.67 g, 4.29 mmol) was suspended in toluene (3.0 mL) and HMDS (0.90 mL). The mixture was heated at 100°C for 3 h. After concentration, the residue was dissolved in dichloromethane (10 mL). The intermediate 7 (1.22 g, 4.68 mmol) and SnCl4 (1.34 g, 5.25 mmol) were added to the mixture at 0°C. The mixture was stirred at 0°C for 2 h. NaHCO3 (2.5 g) and H2O (0.8 mL) was added to the mixture. The mixture was stirred at room temperature for 2 h. Filtration gave a solution, which was washed by 1N NaHCO3. The solvent was evaporated to give a crude product, which was purified by flash column chromatography (1:30 MeOH/CH2Cl2) to give a vacuum dried product as a white solid 10 (0.85 g, 56%), mp: 146 –149°C. 1H NMR (400 MHz, CDCl3): ␦ 1.52–1.54 (d, J ⫽ 7.5 Hz, 3H, 5⬘-Me), 2.09 –2.12 (d, J ⫽ 7.5 Hz, 3H, CH3CO), 2.13–2.16 (d, J ⫽ 7.5 Hz, 3H, CH3CO), 4.37– 4.39 (d, J ⫽ 5.2 Hz, 1H), 4.99 –5.01 (d, J ⫽ 5.2 Hz, 1H), 5.45 (s, 1H), 5.96 –5.98 (d, J ⫽ 7.2 Hz, 1H), 7.53 (broad single peak brs)), 1H), 7.95 (brs, 1H), 8.95– 8.98 (d, J ⫽ 10 Hz, 1H). LRMS (CI, m/z): 379.3 ([M ⫹ Na]⫹, 100%). HRMS (CI, m/z): calcd for C13H 16N4O8Na 379.0866, found 379.0870. 4.3.3. 2⬘,3⬘-Di-O-acetyl-5⬘-deoxy-5-nitro-N4(pentyloxycarbonyl)cytidine (11) The mixture of compound 10 (0.53 g, 1.49 mmol) in dichloromethane (25 mL) and pyridine (3.0 mL) was cooled to 0°C. n-Pentyl chloroformate (0.44 g, 2.88 mmol) was added to the mixture. The mixture was refluxed for 4 h. After evaporation, the crude product was purified by flash column chromatography (1:20 MeOH/CH2Cl2) to give a vacuum dried product as a white solid 11 (0.58 g, 84%), mp: 50°C. 1H NMR (400 MHz, CDCl3): ␦ 1.18 –1.50 (m, 6H), 1.52–1.55 (t, J ⫽ 7.3 Hz, 3H), 2.03–2.05 (t, J ⫽ 7.6 Hz,

1040

Q.-H. Zheng et al. / Nuclear Medicine and Biology 31 (2004) 1033–1041

3H), 2.08 –2.10 (d, J ⫽ 8.4 Hz, 3H), 2.13–2.15 (d, J ⫽ 8.4 Hz, 3H), 4.11– 4.15 (t, J ⫽ 7.8 Hz, 1H), 4.36 – 4.38 (d, J ⫽ 5.6 Hz, 1H), 4.99 –5.01 (d, J ⫽ 5.2 Hz, 1H), 5.50 (s, 1H), 5.93–5.94 (d, J ⫽ 5.2 Hz, 1H), 7.98 (s, 1H), 8.28 (brs, 1H), 8.95– 8.97 (d, J ⫽ 8.0 Hz, 1H). LRMS (CI, m/z): 471.2 ([M ⫹ H]⫹, 11.6%), 201.2 (100%). HRMS (CI, m/z): calcd for C19H27O10N4 471.1727, found 471.1731.

chromatography (1:50 MeOH/CH2Cl2) to give a vacuum dried product as a white solid 1 (65 mg, 82%), mp: 88 – 89°C (see Ref. [9], mp: 119 –121°C). 1H NMR (400 MHz, CDCl3): ␦ 0.89 – 0.93 (t, J ⫽ 6.8 Hz, 3H), 1.26 –1.36 (m, 4H), 1.41–1.43 (d, J ⫽ 6.0 Hz, 3H), 1.68 –1.71 (t, J ⫽ 6.8 Hz, 2H), 3.88 (s, 1H), 4.18 – 4.29 (m, 6H), 5.73 (s, 1H), 7.80 –7.98 (brs, 2H).

4.4. Synthesis of reference standard

4.5. Synthesis of target tracer

4.4.1. 2⬘,3⬘-Di-O-acetyl-5⬘-deoxy-5-fluorocytidine (13) The mixture of 5-fluorocytosine (12, 0.30 g, 2.32 mmol) in toluene (1.5 mL) and HMDS (0.38 g, 2.32 mmol) was refluxed for 3 h. The solvent was removed by evaporation and the residue was dissolved in dichloromethane (5 mL). Compound 7 (0.66 g, 2.54 mmol) and SnCl4 (0.72 g, 0.32 mmol) were added to the solution at 0°C. The mixture was stirred at 0°C for 2 h. NaHCO3 (1.2 g) and H2O (0.5 mL) was added to the mixture. After the mixture was stirred at room temperature for 3 h, filtration gave a solution, which was washed by 1N NaHCO3. The solvent was removed by evaporation to give a crude product, which was purified by flash column chromatography (1:20 MeOH/EtOAc) to give a vacuum dried product as a white solid 13 (0.60 g, 78%), mp: 81– 83°C (see Ref. [9], mp: 191.5–192.3°C). 1H NMR (400 MHz, CDCl3): ␦ 1.42–1.43 (d, J ⫽ 6.4 Hz, 3H, 5⬘-Me), 2.06 (s, 3H, CH3CO), 2.07 (s, 3H, CH3CO), 4.19 – 4.22 (m, 2H), 4.96 – 4.97 (t, J ⫽ 6.6 Hz, 1H), 5.29 –5.30 (m, 1H), 5.96 –5.97 (d, J ⫽ 3.2 Hz, 1H), 7.35–7.37 (dd, J ⫽ 3.2, 5.6 Hz, 1H), 7.90 – 8.10 (brs, 1H).

4.5.1. 5⬘-Deoxy-5-[18F]fluoro-N4(pentyloxycarbonyl)cytidine ([18F]Xeloda; [18F]1) No-carrier-added (NCA) aqueous H18F (0.5 mL) prepared by 18O(p, n)18F nuclear reaction in a RDS-112 cyclotron on an enriched H2 18O water (95 ⫹ %) target was added to a Pyrex vessel which contains K2CO3 (4 mg, in 0.2 mL H2O) and Kryptofix 2.2.2 (10 mg, in 0.5 mL CH3CN). Azeotropic distillation at 115°C with HPLC grade CH3CN (3 ⫻ 1 mL) under a nitrogen steam efficiently removed water to form anhydrous K18F-Kryptofix 2.2.2 complex. The nitro-precursor 11 (2–3 mg, dissolved in 0.5 mL DMSO) was introduced to the K18F-Kryptofix 2.2.2 complex. The reaction mixture was sealed and heated at 150°C for 15–20 min and was subsequently allowed to cool down, at which time the crude product was passed through a Silica Sep-Pak cartridge to remove Kryptofix 2.2.2 and unreacted precursor 11 and K18F. The Sep-Pak column was eluted with 15% MeOH/CH2Cl2 (5.0 mL), and the fractions were passed onto a rotatory evaporator. The solvent was removed by evaporation under high vacuum (0.1–1.0 mmHg) to give a crude product [18F]14. This residue was added with 6 N NaOH (0.1 mL) and MeOH (1.0 mL) and heated for 10 min at 80°C. The contents were neutralized with 2 N HCl (0.3 mL), diluted with ethanol (3 mL), and evaporated under vacuum to give a crude product [18F]1. The contents of the reaction tube were diluted with 3:1:3 CH3CN/MeOH/20 mM, pH 6.7 KHPO⫺ 4 (1.5 mL), and injected onto the preparative HPLC column. The product fraction was collected, the solvent was removed by rotatory evaporation under vacuum, and the final product [18F]1 was formulated in saline, sterile-filtered through a sterile vented Millex-GS 0.22 ␮m cellulose acetate membrane and collected into a sterile vial. The radiochemical yield of [18F]1 was 20 –30%, and the synthesis time was 60 –70 min from EOB. Retention times in the analytical HPLC system were: Rt11 ⫽ 2.38 min, Rt[18F]14 ⫽ 4.02 min, Rt[18F]1 ⫽ 3.19 min, RtK18F ⫽ 1.98 min. Retention times in the preparative HPLC system were: Rt[18F]14 ⫽ 6.54 min, Rt[18F]1 ⫽ 4.58 min.

4.4.2. 2⬘,3⬘-Di-O-acetyl-5⬘-deoxy-5-fluoro-N4(pentyloxycarbonyl)cytidine (14) The mixture of compound 13 (0.16 g, 0.49 mmol) in dichloromethane (25 mL) and pyridine (2 mL) was cooled to 0°C and n-pentyl chloromate (0.11 g, 0.73 mmol) was added to the mixture. The mixture was stirred at 0°C for 1 h. After evaporation, the crude product was purified by flash column chromatography (1:50 MeOH/CH2Cl2) to give a vacuum dried product as a light yellow solid 14 (0.16 g, 73%), mp: 50 –51°C. 1H NMR (400 MHz, CDCl3): ␦ 0.91– 0.92 (d, J ⫽ 6.0 Hz, 3H), 1.24 –1.58 (m, 6H), 1.71–1.73 (d, J ⫽ 6.4 Hz, 3H), 2.10 (s, 3H, CH3CO), 2.12 (s, 3H, CH 3CO), 4.13– 4.19 (m, 2H), 4.24 – 4.28 (t, J ⫽ 6.4 Hz, 2H), 5.01–5.03 (d, J ⫽ 5.6 Hz, 1H), 5.29 –5.30 (d, J ⫽ 4.8 Hz, 1H), 5.96 (s, 1H), 7.40 (s, 1H). 4.4.3. 5⬘-Deoxy-5-fluoro-N4-(pentyloxycarbonyl)cytidine (Xeloda; Capecitabine; 1) Compound 14 (97 mg, 0.22 mmol) was dissolved in methanol (2 mL). NaOH (35 mg, 0.88 mmol) in H2O (0.5 mL) was added to the solution. The mixture was stirred at 0°C for 0.5 h. 2N HCl was added to adjust pH to 7. The organic solvent was removed and the residue was extracted by EtOAc. The organic phase was dried over Na2SO4. After evaporation, the crude product was purified by flash column

Acknowledgments This work was partially supported by the Susan G. Komen Breast Cancer Foundation Grant No. IMG 02-1550 (to QHZ), the National Institutes of Health/National Cancer Institute Grant No. P20CA86350 (to GDH), the Indiana 21st

Q.-H. Zheng et al. / Nuclear Medicine and Biology 31 (2004) 1033–1041

Century Research and Technology Fund (to GDH), and the Lilly Endowment Inc. (to the Indiana Genomics Initiative (INGEN) of Indiana University). The authors thank Barbara Glick-Wilson and Michael Sullivan for the efforts in producing [18F]fluoride. The referee’s criticisms and editor’s comments for the revision of the manuscript are greatly appreciated.

[15]

[16]

References [1] McDonald F, Miles D. Xeloda and Taxotere: a review of the development of the combination for use in metastatic breast cancer. Int J Clin Pract 2003;57:530 – 4. [2] Hoff PM. Practical considerations in the use of oral fluoropyrimidines. Semin Oncol 2003;30(3 Suppl 6):88 –92. [3] Wenzel C, Mader RM, Steger GG, Pluschnig U, Kornek GV, Scheithauer W, Locker GJ. Capecitabine treatment results in increased mean corpuscular volume of red blood cells in patients with advanced solid malignancies. Anticancer Drugs 2003;14:119 –23. [4] Fujiwaki R, Hata K, Nakayama K, Moriyama M, Iwanari O, Katabuchi H, Okamura H, Sakai E, Miyazaki K. Thymidine kinase in epithelial ovarian cancer: relationship with the other pyrimidine pathway enzymes. Int J Cancer 2002;99:328 –35. [5] Pentheroudakis G, Twelves C. The rational development of capecitabine from the laboratory to the clinic. Anticancer Res 2002;22: 3589 –96. [6] McGavin JK, Goa KL. Capecitabine: a review of its use in the treatment of advanced or metastatic colorectal cancer. Drugs 2001; 61:2309 –26. [7] Kuhn JG. Fluorouracil and the new oral fluorinated pyrimidines. Ann Pharmacother 2001;35:217–27. [8] Mrozek-Orlowski ME, Frye DK, Sanborn HM. Capecitabine: nursing implications of a new oral chemotherapeutic agent. Oncol Nurs Forum 1999;26:753– 62. [9] Shimma N, Umeda I, Arasaki M, Murasaki C, Masubuchi K, Kohchi Y, Miwa M, Ura M, Sawada N, Tahara H, Kuruma I, Horii I, Ishitsuka H. The design and synthesis of a new tumor-selective fluoropyrimidine carbamate, capecitabine. Bioorg Med Chem 2000; 8:1697–706. [10] Hattori K, Kohchi Y, Oikawa N, Suda H, Ura M, Ishikawa T, Miwa M, Endoh M, Eda H, Tanimura H, Kawashima A, Horii I, Ishitsuka H, Shimma N. Design and synthesis of the tumor-activated prodrug of dihydropyrimidine dehydrogenase (DPD) inhibitor, RO0094889 for combination therapy with capecitabine. Bioorg Med Chem Lett 2003; 13:867–72. [11] Weissleder R, Mahmood U. Molecular imaging. Radiology 2001; 219:316 –33. [12] Toyohara J, Fujibayashi Y. Trends in nucleoside tracers for PET imaging of cell proliferation. Nucl Med Biol 2003;30:681–5. [13] Kiss J, Dsouza R, VanKoeveringe JA, Arnold W. Stereospecific synthesis of the anticancer agent 5-deoxy-5-fluoroyurdine (5-DFUR) and its 5-deuteriated derivatives. Helv Chim Acta 1982;65:1522–37. [14] Wang J-Q, Zheng Q-H, Fei X, Mock BH, Hutchins GD. Novel radiosynthesis of PET HSV-tk gene reporter probes [18F]FHPG and

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

1041

[18F]FHBG employing dual Sep-Pak SPE techniques. Bioorg Med Chem Lett 2003;13:3933– 8. Zheng Q-H, Wang J-Q, Liu X, Fei X, Mock BH, Glick-Wilson BE, Sullivan ML, Raikwar SP, Gardner TA, Kao C, Hutchins GD. An improved total synthesis of PET HSV-tk gene reporter probe 9-(4[18F]fluoro-3-hydroxymethylbutyl)guanine ([18F]FHBG). Syn Commun 2004;34(4):689 –704. Wang J-Q, Zheng Q-H, Fei X, Liu X, Gardner TA, Kao C, Raikwar S, Glick-Wilson BE, Sullivan ML, Mock BH, Hutchins GD. An improved total synthesis of PET HSV-tk gene expression imaging agent 9-[(3-[18F]fluoro-1-hydroxy-2-propoxy)methyl]guanine ([18F]FHPG). Syn Commun 2004;34(5):917–32. Zheng Q-H, Fei X, Liu X, Wang J-Q, Sun HB, Mock BH, Stone KL, Martinez TD, Miller KD, Sledge GW, Hutchins GD. Synthesis and preliminary biological evaluation of MMP inhibitor radiotracers [11C]methyl-halo-CGS 27023A analogs, new potential PET breast cancer imaging agents. Nucl Med Biol 2002;29:761–70. Zheng Q-H, Liu X, Fei X, Wang J-Q, Ohannesian DW, Erickson LC, Stone KL, Hutchins GD. Synthesis and preliminary biological evaluation of radiolabeled O6-benzylguanine derivatives, new potential PET imaging agents for the DNA repair protein O6-alkylguanineDNA alkyltransferase in breast cancer. Nucl Med Biol 2003;30:405– 15. Zheng Q-H, Fei X, DeGrado TR, Wang J-Q, Stone KL, Martinez TD, Gay DJ, Baity WL, Mock BH, Glick-Wilson BE, Sullivan ML, Miller KD, Sledge GW, Hutchins GD. Synthesis, biodistribution and microPET imaging of a potential cancer biomarker carbon-11 labeled MMP inhibitor (2R)-2-[[4-(6-fluorohex-1-ynyl)phenyl]sulfonylamino]-3methylbutyric acid [11C]methyl ester. Nucl Med Biol 2003;30:753– 60. Zheng Q-H, Stone KL, Mock BH, Miller KD, Fei X, Liu X, Wang J-Q, Glick-Wilson BE, Sledge GW, Hutchins GD. [11C]Choline as a potential PET marker for imaging of breast cancer athymic mice. Nucl Med Biol 2002;29:803–7. Zheng Q-H, Liu X, Fei X, Wang J-Q, Mock BH, Glick-Wilson BE, Sullivan ML, Hutchins GD. Facile synthesis of [11C]edrophonium and its analogs as new potential PET imaging agents for heart acetylcholinesterase. Bioorg Med Chem Lett 2003;13:1787–90. Mulholland GK, Zheng Q-H, Mock BH, Vavrek MT. Convenient labeling and isolation of [11C-methyl]quaternary amines. J Label Compd Radiopharm 1999;42(Suppl. 1):S459 – 61. Mock BH, Winkle W, Vavrek MT. A color spot test for the detection of Kryptofix 2.2.2 in [18F]FDG preparations. Nucl Med Biol 1997; 24:193–5. Yun M, Oh SJ, Ha HJ, Ryu JS, Moon DH. High radiochemical yield synthesis of 3⬘-deoxy-3⬘-[18F]fluorothymidine using (5⬘-O-dimethoxytrityl-2⬘-deoxy-3⬘-O-nosyl-␤-D-threo-pentofuranosyl)thymine and its 3-N-BOC-protected analogue as a labeling precursor. Nucl Med Biol 2003;30:151–7. Fei X, Zheng Q-H, Liu X, Wang J-Q, Stone KL, Miller KD, Sledge GW, Hutchins GD. Synthesis of MMP inhibitor radiotracer [11C]CGS 25966, a new potential PET tumor imaging agent. J Label Compd Radiopharm 2003;46:343–51. Zheng Q-H, Mulholland GK. Isomerization study of aryltropane analogs of cocaine. Syn Commun 2000;30:333– 40.