Automated synthesis and purification of [18F]bromofluoromethane at high specific radioactivity

https://doi.org/10.1016/S0969-8043(00)00358-4Get rights and content

Abstract

[18F]Bromofluoromethane was synthesised from dibromomethane by substitution of bromine with [18F]fluoride. The synthesis and separation of the [18F]bromofluoromethane were automated. [18F]Bromofluoromethane was used to convert a phenolic and a thiophenolic precursor into a labelled ether and thioether, respectively. The specific radioactivity of these labelled products was determined with both high-performance liquid chromatography (with UV-absorbance detection) and liquid chromatography (with mass spectrometric detection). The median for the specific radioactivity, corrected at the end of radionuclide production, was 934 GBq/μmol (range 40–9900 GBq/μmol; n=83).

Introduction

Although fluorine is foreign to most biological molecules, it has found widespread use in pharmaceuticals. Fluorine introduction can have a profound effect on the lipophilicity of small molecules (Park and Kitteringham, 1994), and is widely used to modify drug action. The fluorine atom is about the same size as the hydrogen atom. Therefore, fluorine has the quality to mimic hydrogen atoms in organic molecules with respect to steric requirements.

Fluorine-18 (18F; β+=96.7%, T1/2=109.8min) is of considerable importance in radiochemistry for positron emission tomography (PET) due to its near optimal decay characteristics. Numerous methods for introducing 18F into organic molecules have been developed (Kilbourn, 1990; Fowler and Wolf, 1997). A novel way of labelling organic molecules with 18F is by using 18F-labelled fluorohalomethane as an alkylation reagent. The radiochemistry of 11C-alkylations (11C; β+=99.8%, T1/2=20.4min) is well developed for PET, as one of the first and nowadays most common synthetic precursors is [11C]methyl iodide (Comar et al., 1976; Långström and Lundqvist, 1976). However, in many instances, the short half-life of 11C limits its usefulness in dynamic PET experiments to about 100 min (Fowler and Wolf, 1997). Another aspect is the specific radioactivity (SA) of 11C versus 18F. In theory, due to shorter half-life, the maximum SA for 11C is higher than for 18F (341 versus 63.4 TBq/μmol, respectively). In practice, it turns out to be quite difficult to prepare 11C-methylation agents with SA greater than 700 GBq/μmol, whereas careful exclusion of carrier fluoride in 18F production systems can give SA in excess of 4 TBq/μmol. This difference is accentuated by the fact that SA will decline with the half-life of the radioisotope.

So far there have not been many studies reported on alkylation with the [18F]fluoromethyl group. Coenen et al. (1985) synthesised [18F]bromofluoromethane but did not elaborate on its usefulness in labelling chemistry. Zheng and Berridge (1997), Zheng and Berridge (2000) synthesised [18F]fluoroiodomethane, which was isolated by simple distillation and used for the labelling of model compounds.

Our aim was to develop an automated unit for the production and separation of radiofluoromethylation agents for the large activity production of radiopharmaceuticals for PET. After initial development work, [18F]bromofluoromethane was chosen as the labelling agent. Here we describe the automated synthesis and purification of [18F]fluorobromomethane at high SA, and demonstration of its use as a labelling agent.

Section snippets

Materials

If not otherwise stated, chemicals were acquired from Merck AG (Darmstadt, Germany) or Fluka Chemie (Buchs, Switzerland).

Radionuclide production

[18F]Fluoride was made by irradiating [18O]water (>94 at%; 400 or 700 μl; Isotec Inc., Miamisburgh, OH) in silver target chambers (Solin et al., 1988) with 17 MeV protons from a 103 cm AVF cyclotron. Typical irradiations were of 45 min duration with a beam current of 10 μA yielding about 18 GBq [18F] fluoride. After irradiation, the target water was transported via either silicone

Yields of [18F] fluorohalomethanes

The decay-corrected yield of [18F]fluoroiodomethane from [18F]fluoride was 5.7±5.5% (n=30). The maximum yield was 20.8%, based on the radioactivity trapped in vessel B. The radiochemical purity was higher than 95% as determined by HPLC. However, because the product was not purified by GC, various amounts of unidentified non-radioactive impurities were also transferred to vessel B. Because of the low and varying yields further efforts to improve this synthesis were abandoned.

The yield of [18

Discussion

An automated apparatus was developed for the [18F]fluoromethylation of compounds with [18F]fluorohalomethanes to provide prospective radiopharmaceuticals. Both [18F]fluoroiodomethane and [18F]bromofluoromethane were initially considered as labelling agents. Dihalogenated methanes were used as starting materials in the synthesis of these fluorohalomethanes. It was considered important to remove these materials completely from the labelled fluorohalomethanes, as they will react and consume the

Conclusions

In the present study we designed a method for producing [18F]bromofluoromethane which could subsequently be successfully used in fluoroalkylation reactions. The labelling reactions showed good repeatability and yields, and an automated synthesis apparatus was constructed. Although relatively high, the SA showed considerable variation. The 19F sources causing the SA variation have not yet been identified. LC-MS will be used to measure the mass ratio of 18F to 19F in these high SA products and

Acknowledgements

Professor Sharon Stone-Elander, Ph.D., is gratefully acknowledged for her help and advice. We thank Esa Kokkomäki, for his assistance with computers and automation. Stefan Johansson, Erkki Stenvall and Per-Olof Eriksson at the Turku PET Centre Accelerator Laboratory are also kindly acknowledged.

References (17)

There are more references available in the full text version of this article.

Cited by (0)

View full text