Synthesis of 1,3,4,6-Tetra-O-acetyl-2-O-trifluoromethanesulfonyl-β-D-mannopyranose (Mannose Triflate) Precursor for the Production of [¹⁸F] FDG

Abstract

Purpose of the Study:

1,3,4,6-tetra-O-acetyl-2-O-trifluoromethanesulfonyl-β-D-mannopyranose (mannose triflate), the precursor used for the synthesis of [¹⁸F] Fluorodeoxyglucose ([¹⁸F] FDG) is imported from a few commercial suppliers abroad. As part of self-reliance, a reliable synthesis and characterization of mannose triflate has been developed, details of which are reported in this paper.

Materials and Methods:

Synthesis of 1,3,4,6-tetra-O-acetyl-2-O-trifluoromethanesulfonyl-β-D-mannopyranose (Mannose triflate) carried by Triflation of 1,3,4,6-Tetra-O-acetyl-β-D-mannopyranose with Tf2O–pyridine under argon atmosphere for 6 h. Synthesized mannose triflate is used for the production of [¹⁸F] FDG using siemens explora FDG-4-automated module.

Results:

Starting from 2.6 g of 1,3,4,6-Tetra-O-acetyl-β-D-mannopyranose, mannose triflate was synthesized with an overall yield of ~ 3.0 g (~80%). The synthesis was systematically scaled up to 52 g based on the experience from the initial small size batches. The characterization of the product was done by Infra Red, Nuclear Magnetic Resonance and Mass spectroscopy. The precursor was used in the manufacture of [¹⁸F] FDG using Siemens Explora synthesis module. Radiochemical yields of 55% ± 2% (decay uncorrected), radiochemical purity >96% were obtained which is comparable with the imported precursor. All the quality control parameters were within the limits as per pharmacopoeia when the in-house synthesized mannose triflate was used.

Conclusion:

A large-scale preparation of mannose triflate is now carried out to satisfy the growing needs for FDG in positron emission tomography centers and hospitals.

Introduction

Positron emission tomography-computed tomography (PET-CT) imaging with [¹⁸F] Fluorodeoxyglucose ([¹⁸F] FDG) is routinely used for the accurate staging as well as for evaluation of therapy response and recurrence of many types of cancers.[1] The glycolytic pathway uptake mechanism of FDG in the human body is well documented; high glucose metabolism is responsible for the higher uptake of FDG.[2] Glucose taken up by the cells undergoes a series of chemical reactions starting with the conversion to glucose-6-phosphate. [¹⁸F] FDG also gets converted to [¹⁸F] FDG-6-phosphate and all further reactions are stopped thereby accumulation of the radiopharmaceutical in the cells.[3] Cancer cells have high metabolism and hence higher uptake of [¹⁸F] FDG as compared to normal cells. The accumulation of [¹⁸F] FDG is mapped using a PET-CT imaging which gives valuable information for patient management.[4]

[¹⁸F] FDG is synthesized in automated modules through nucleophilic synthesis using ¹⁸F-ions obtained by irradiation of ¹⁸O enriched water.[5] In India, there are 24 cyclotrons engaged in the production of [¹⁸F] FDG to be utilized in nearly 400 PET-CT machines spread across the country. Mannose triflate is the precursor for the synthesis of [¹⁸F] FDG and according to bibliographic analysis, there are a few reports have been published on its synthesis. Hamacher developed a method in 1984 to create mannose triflate which involved selective acylation of D-mannose to give 1,3,4,6-tetra-O-acetyl-β-D-mannopyranose followed by the addition of triflate group to get mannose triflate.[6] Pavliak and Kovác modified the synthetic route giving higher yield of ~90% for mannose triflate synthesis.[7] In 1992, Pozsgaya et al., described the synthesis of mannose triflate (yield 65%–70%) from 1,2,3,4,6-penta-O-acetyl-α,β-D-mannopyranose as a midway step in the conversion of methyl-2-azido-2-deoxy-l-thio-a-D-glucopyranoside to β-D-glucopyranoside.[8] Toyokuni et al. used D-mannose as the starting material, followed by acylation and substitution of a triflate group in the second position of 1,3,4,6-tetra-O-acetyl-β-D-mannopyranose via a five-step process with 75%–85% yield and utilized it for FDG synthesis.[9,10,11]

There is a total dependence on import for the reagents and kits used for the manufacture of [¹⁸F] FDG in India. Hence, there exists a need for the development of indigenous technology towards synthesis of reagents as well as kits for ¹⁸F radiopharmaceuticals production. As a step towards indigenization, we have done in-house production of mannose triflate using 1,3,4,6-tetra-O-acetyl-β-D-mannopyranose and triflic anhydride under mild conditions and used it in our production procedure as represented in Figure 1. [¹⁸F] FDG synthesis had yields comparable to imported precursor and the product confirmed to purity standards as per pharmacopoeia. The details of synthesis of mannose triflate are documented in this paper.

Figure 1.

Synthesis scheme of mannose triflate

Materials and Methods

Materials

1,3,4,6-Tetra-O-acetyl-β-D-mannopyranose and trifluoro methane sulfonic anhydride were purchased from Tokyo Chemical Company Industry (TCI), Japan. Anhydrous pyridine and dichloromethane were purchased from Merck. All the solvents were used without further purification.

¹H NMR spectra were recorded either on a Bruker Avance 400 MHz spectrometer or Bruker Avance 500 MHz spectrometer. CDCl₃ was used as solvent and TMS as internal standard for ¹H NMR spectra. ¹⁹F NMR spectra were recorded in Bruker Avance 500 MHz spectrometer (470 MHz for ¹⁹F). CDCl₃ was used as solvent and CFCl₃ as an internal standard for ¹⁹F NMR spectra. Quantitative NMR (qNMR) method was used to determine the purity of the samples. Liquid chromatography-Electrospray Ionization-High Resolution Mass Spectroscopy (HRMS) spectra were recorded on Thermo Fisher Scientific, Q Exactive mass spectrometer. Melting point was measured on an Environmental and Scientific Instruments (EI) Digital Melting Point Apparatus 934. Thin-layer chromatography (TLC) analyses were performed on precoated TLC silica gel 60 F254 (Merck) plates using Ethylacetate: Petroleum Ether (1:1) as the eluent and the spots were visualized by spraying iodine vapors.

Siemens Eclipse HP 11 MeV cyclotron was used for the production of ¹⁸F and Siemens Explora FDG-4 module was used for the synthesis of [¹⁸]FDG. HPLC-grade chemicals, such as acetonitrile and HCl, were procured from Merck. Krptofix 222 was procured from ABX. C-18 and alumina cartridges were procured from Waters Corporation. AG-11 and AG-50 resins were procured from Bio-Rad Laboratories, USA. Agilent gas chromatography was used for the analysis of residual solvents in the [¹⁸F] FDG. TLC scanner with radiation detector supplied by Eckert and Ziegler was used for quantification of radioactivity in the TLC strip.

Methods

Synthesis of 1,3,4,6-tetra-O-acetyl-2-O-trifluoromethanesulfonyl-β-D-mannopyranose

The mannose triflate synthetic procedure was adapted from the reported work by Toyokuni et al.[6] The initial synthetic step of acetylation of D-mannose was excluded due to the easy availability of 1,3,4,6-tetra-O-acetyl-β-D-mannopyranose.[7,8,11]

1,3,4,6-Tetra-O-acetyl-β-D-mannopyranose 1 (2.6 g, 7 mmol) was dissolved in anhydrous CH₂Cl₂ (20 mL) containing dry pyridine (1.5 mL, 0.186 mol) in a 250 mL RB flask. The mixture was cooled to −15°C in an ice bath continuing salt (NaCl: Ice [3:1] bath). Trifluoromethanesulfonic anhydride (2.2 mL, 13 mmol) was added dropwise over a period of 40 min under dry argon atmosphere with vigorous stirring. The mixture was slowly allowed to reach room temperature which took about 6 h. The reaction mixture was successively washed with ice-cold saturated aqueous NaHCO₃ (40 mL) and water (40–50 mL). The organic layer was dried over anhydrous Na₂SO₄, filtered, and concentrated on a rotary evaporator at 30°C. The resultant solid residue was recrystallized from absolute ethanol to give the mannose triflate as white needles.(yield: 2.86 g, 76%): Melting Point: 120°C (lit. 118°C–120°C); IR (ν_max): 1753, 1730, 1414, 1215 cm⁻¹. ¹H NMR (400 MHz, CDCl_{3 ppm}) δ: 5.91 (d, J = 0.9 Hz, 1H), 5.30 (t, J = 9.9 Hz, 1H), 5.19 (dd, J = 10.0, 3.0 Hz, 1H), 5.15 (dd, J = 3.1, 0.9 Hz, 1H), 4.22 (qd, J = 12.5, 3.8 Hz, 2H), 3.84 (ddd, J = 9.9, 5.2, 2.5 Hz, 1H), 2.17 (s, 3H), 2.15–2.04 (m, 9H). ¹⁹F NMR (471 MHz, CDCl_3, ppm) δ: −73.97, (CF₃). HRMS calculated for C₁₅H₁₉F₃O₁₂S (M + Na) +: 503.0442, found 503.0440.

The batch size was scaled up to 52 g of mannose triflate as follows: we used 1,3,4,6-Tetra-O-acetyl-β-D-mannopyranose 49.3 g (0.144 mol), triflic anhydride 27 mL (0.16 mmol), and pyridine 25 mL (0.36 mmol) in dry DCM. We followed the same procedure and obtained 52 g of mannose triflate.

Production of flourine-18

The preloaded tantalum targets with 5.2 mL of [¹⁸O] H₂O having ¹⁸O enrichment >98% was irradiated with 11 MeV protons having beam current of 118–120 µA. The duration of irradiation was 180–200 min. After irradiation, [¹⁸O] H₂O containing ¹⁸F^-was transferred to the Siemens Explora FDG-4 synthesis module for the production of [¹⁸F] FDG.

Automated synthesis of fluodeoxyglucose

The ¹⁸F^-ions in the ¹⁸O-enriched water is passed through the quaternary methyl ammonium (QMA) cartridge. QMA is a strong anion exchanger and ¹⁸F^-ions are trapped efficiently. The irradiated ¹⁸O enriched which contain residual ¹⁸F^-ions as well as some other radioactive impurities, are collected in a recovery vial kept inside the hot cell. ¹⁸F^-ions absorbed in the QMA cartridge are eluted into the reaction vial with 1.8 mL of 0.0016 M of kryptofix prepared by dissolving 600 mg of kryptofix in 19 mL of acetonitrile mixed with 0.0008 M K₂CO₃ (110 mg) dissolved in 1 mL of water. This mixture is heated for 5 min at 95°C. Further, 1 mL of acetonitrile is added to the reaction mixture and heated at 95°C azeotropically for 5 min to remove water and make the reaction mixture completely dry. One milliliter of 0.001 M mannose triflate, which is prepared by dissolving 500 mg of mannose triflate in 20 mL of acetonitrile was added to the reaction mixture and heated for 3–5 min at 95°C. The radiolabelling of the mannose triflate precursor is achieved at this stage and hydrolysis of the labeled product is done by adding 2 mL of 1M HCl to the reaction mixture and heating for 8 min at 120°C. The reaction scheme is represented in Figure 2.

Figure 2.

Reaction scheme of fluodeoxyglucose synthesis. [¹⁸F] FDG: [¹⁸F] fluodeoxyglucose

The reaction mixture underwent a series of purification steps. An in-house prepared purification column containing cation exchange resin (AG^®50W-X8) and ion retardation resin (AG^®11-A8) was used in the first step of purification. Cation exchange resin is used to remove kryptofix while ion retardation resin is used to remove excess acid and neutralize the FDG. A C-18 cartridge in series was to remove any unhydrolyzed FDG which was followed by passing through an alumina cartridge to remove free fluoride ions. The purified product was passed through a 0.22 µ filter to the product vial kept in the dispensing hot cell. The product activity was measured in a Capintec dose calibrator. The volume of [¹⁸F] FDG ranged from 11 to 18 mL in different batches which was diluted as needed for dispensing into individual customer requirement.

Quality control of [¹⁸F] fluodeoxyglucose

The radiochemical purity of the FDG was checked by ITLC-SG using Radio-TLC scanner. Acetonitrile with 5% water was used as the mobile phase. The concentration of kryptofix was estimated by the iodine spot test. The pH of the product was tested with pH paper. The residual solvents, such as acetonitrile and ethanol were analyzed using gas chromatography. The bacterial endotoxin was quantified using a Charles River BET device. The sample for BET was prepared in 1:200 dilution of FDG sample with LAL water and 25 µL aliquots of the sample were dispensed into four wells of precalibrated BET Charles River cartridges and was inserted into the BET device. The final endotoxin unit value was observed on the BET device’s screen after 10–15 min. The sterility of the FDG solution was tested using tryptone soya broth (TSB) and fluid thioglycolate (FTG) media. One milliliter of diluted FDG sample was injected into the media. TSB media was incubated at 22°C ± 2.5°C and FTG media at 32°C ± 2.5°C. Both the inoculated media were observed for turbidity up to 14 days. The absence of turbidity indicates that the product is sterile.

Results and Discussion

The synthesis of mannose triflate was successfully achieved through the reaction of 1,3,4,6-Tetra-O-acetyl-β-D-mannopyranose with trifluoromethanesulfonic anhydride in anhydrous CH₂Cl₂. The reaction proceeded smoothly under a dry argon atmosphere with vigorous stirring, and the resulting mannose triflate was obtained as white needles after recrystallization from absolute ethanol. The complete re-crystallized product can be recovered within 3 days. The yield of the synthesis was found to be 80% in five batches, indicating a high efficiency in the conversion of the starting material to the desired product.

The product, mannose triflate, was characterized by various analytical techniques. In the IR spectrum [Supplementary data S1 ^{(1.5MB, tif)} ] of 1,3,4,6-Tetra-O-acetyl-β-D-mannopyranose, the peak at 3500 cm⁻¹ corresponding to the OH group was absent in mannose triflate, indicating quantitative conversion of the OH group into the triflate group resulting in high yields of mannose triflate. The IR spectrum exhibited characteristic peaks at 1753 cm⁻¹ and 1730 cm⁻¹, indicating the presence of carbonyl functionalities in mannose triflate. Additional peaks at 1414 cm⁻¹ and 1215 cm⁻¹ further supported the successful formation of the triflate derivative.

The ¹H NMR spectrum [Supplementary data S2 ^{(1.5MB, tif)} ] showed distinct signals, confirming the structure of mannose triflate. ¹⁹F NMR spectrum [Supplementary data S3 ^{(862.4KB, tif)} ] displayed a single peak at δ −73.94, corresponding to the trifluoromethyl group (CF₃) of the triflate moiety.[10] HRMS [Supplementary data S4 ^{(1.3MB, tif)} ] analysis confirmed the molecular weight of the product, with the calculated mass for C₁₅H₁₉F₃O₁₂S + Na⁺ (M + Na) + at 503.0442 and the observed mass at 503.0440. This further validated the successful synthesis of the desired mannose triflate. The synthesis of 1,3,4,6-tetra-O-acetyl-2-O-trifluoromethanesulfonyl-β-D-mannopyranose was efficiently accomplished, resulting in a high yield of 80% in different batches. The purity was 98% and above as per qNMR. The comprehensive characterization using IR, ¹H NMR, ¹⁹F NMR, and HRMS confirmed the successful formation of the triflate derivative.

Scaling up of the reaction was done up to 52 g by enhancing the volume of the reactants correspondingly. Controlled addition of triflic anhydride is crucial and can simplify the workup of the reaction. The crystalline mannose triflate precursor obtained can be stored in a dark glass vial and desiccated at −20°C and is stable for several months.

Radiochemical yield and radiochemical purity of the [¹⁸F] FDG synthesized was consistent with imported mannose triflate. Radiosynthesis yields were not reduced when this precursor was stored at room temperature and used for about 1 month. In-house prepared mannose triflate is now incorporated in the routine production of [¹⁸F] FDG. The radiochemical yields of all batches are satisfactory. The results of the first six batches of [¹⁸F] FDG synthesized using the in-house mannose triflate is summarized in Table 1. The decay-uncorrected radiochemical yields are 55% ± 2%. The minimum radiochemical yield observed was 51%, with FDG activity 4700 mCi from 8833 mCi of ¹⁸F. The maximum radiochemical yield observed was 57.6% with 4900 mCi of FDG from 8500 mCi ¹⁸F. The radiochemical yields (decay corrected to the end of the bombardment) are 76% ± 3%. These results are consistent with the use of imported mannose triflate. The results of the six batches were perused for introducing the in-house synthesized product to routine production under GMP conditions. The radioTLC pattern of [¹⁸F] FDG is shown in Figure 3. The R_f value of [¹⁸F] FDG is 0.4–0.5 whereas free ¹⁸F and acetylated FDG have the R_f of 0.0 and 0.8–0.9, respectively. We did not observe these peaks in any of the batches of [¹⁸F] FDG produced using in-house prepared mannose triflate. The results of the quality control parameters such as appearance, pH, chemical impurities, residual solvents, bacterial endotoxin test, and sterility are mentioned in the Table 2. The finished product confirmed to the pharmacopoeia specifications.

Table 1.

Results of [¹⁸F] fluodeoxyglucose radiosynthesis using in-house produced mannose triflate

Batch number		18F (mCi@)		[18F] FDG (mCi)		Decay uncorrected radiochemical yield (%)		Decay corrected [18F] FDG activity to EOB		Decay corrected yield (%)
1		8833		4700		53.2		6440		73
2		8700		4450		51.1		6098		70
3		8500		4900		57.6		6714		79
4		8450		4830		57.2		6618		78
5		9066		5050		55.7		6920		76
6		8711		4860		55.8		6659		76

^@The unit of radioactivity used is in mCi as nuclear medicine centers are continuing to use curie unit instead of Becquerel for all practices. FDG: Fluorodeoxyglucose, EOB: End of the Bombardment

Figure 3.

Radio thin-layer chromatography pattern of [¹⁸F] fluodeoxyglucose. EGM-TLC: Electronic gamma measurement- Thin layer chromatography

Table 2.

Quality control results of [¹⁸F] fluodeoxyglucose

Test		Result		Method		Specification
Appearance		Clear		Visual inspection		Clear, colourless or pale yellow solution
pH		5.5		pH paper		4.5–7
Radiochemical purity		>96		ITLC-SG		>95%
Concentration of kryptofix		<10 µg		Spot test		50 µg
Concentration of ethanol		0.01		Gas chromatography		<0.5%
Concentration of ACN		0.001		Gas chromatography		<0.04%
Bacterial endotoxins		<2 EU/mL		Charles river BET		175 EU/volume
Sterility		Sterile		Fluid thioglycolate and tryptic soya broth media test		Sterile
Half life		110		Dose calibrator		109±2 min

ACN: Acetonitrile, BET: Bacterial endotoxin test, ITLC-SG -Instant thin-layer chromatography with silica gel

Conclusion

The first medical cyclotron and PET imaging in India was commissioned in 2002 at the Radiation Medicine Centre, Bhabha Atomic Research Centre. There has been a phenomenon growth in the installation of PET-CT machines in India since then thanks to the availability of PET radiopharmaceuticals mainly [¹⁸F] FDG from private cyclotron centers. As of 2024, there are 24 cyclotrons in operation in India and another four are under installation. It is estimated that there are more than 400 odd PET-CT machines installed in India majority of them are getting the radiopharmaceuticals from the private cyclotron centers. The PET radiopharmaceuticals production in India totally depends on imports. Cyclotron, hot cells, chemistry modules, reagents, kits and ¹⁸O enriched water are sourced from abroad. There is an urgent need to start local developments in this important area to achieve “Atmanirbhar Bharat.”

The Molecular Cyclotron, Cochin uses four synthesis modules of which three are cassette based. The fourth module, Siemens Explora FDG4 module is not cassette based. All reagents are formulated and used. The main advantage of using this module is the reduction in the expense for each batch production. Mannose triflate is the precursor for FDG synthesis and the same is obtained from ABX Advanced Biochemicals Compound, Germany. As part of our efforts towards in-house preparation, we set up an organic synthesis laboratory to synthesize ligands and precursors for radiopharmaceuticals production. Synthesis of mannose triflate was part of this effort. The synthesis procedure was scaled up to 50 g of mannose triflate per batch which is our annual requirement. The synthesis can be further scaled up if needed. This paper is submitted for publication to highlight the need for indigenous development in PET radiopharmaceuticals production as a national initiative and to document our synthesis procedure which can be adapted by others.

Conflicts of interest

There are no conflicts of interest.

Supplementary data

S1.

IR spectrum of Manose triflate (2)

S2.

¹H-NMR spectrum of Manose triflate (2)

S3.

¹⁹F- NMR spectrum of Manose triflate (2)

S4.

LC-HRMS(ESI) spectrum of Manose triflate 2 [503.04-C₁₅H₁₉F₃O₁₂S+Na⁺ (M+Na)⁺; 519.07- C₁₅H₁₉F₃O₁₂S+K⁺ (M+K)⁺]

Funding Statement

Nil.