¹⁸F-FDG Radiosynthesis: A Landmark in the History of PET (perspective on “Efficient Stereospecific Synthesis of No-Carrier-Added 2-[¹⁸F]fluoro-2-Deoxy-D-Glucose Using Aminopolyether Supported Nucleophilic Substitution” J Nucl Med. 1986;27:235–238)

Virtually all of the brain’s energy derives from glucose metabolism. This underpins the development of ¹⁸F-FDG, a radiotracer that was first used to measure regional brain glucose utilization in humans, and has had a profound influence on research in the neurosciences (1). The subsequent discovery that ¹⁸F-FDG can be used to assess viable myocardium and also accumulates in tumors in proportion to their degree of malignancy underpins the evolution of PET as a major clinical tool in cancer diagnosis and monitoring of treatment.

The 1986 landmark paper by Hamacher and colleagues in JNM represents a major milestone in the present use of ¹⁸F-FDG in clinical nuclear medicine worldwide (2).

The ¹⁸F-FDG molecule was modeled on the ¹⁴C-labeled 2-deoxyglucose method, which measures regional brain glucose utilization in animals (3). 2-deoxyglucose is an analog of glucose in which the hydroxyl group on C-2 is replaced by a hydrogen atom. It mimics glucose in serving as a substrate for hexokinase, the rate-limiting step in glycolysis, but does not undergo further steps in the conversion of glucose to energy. Its translation to humans required the development of a radiolabeled version of the 2-deoxy-d-glucose molecule that maintained its biochemical properties and could be labeled with a radioisotope suitable for external imaging in humans. A survey of the literature at the time revealed that 2-deoxy-2-fluoro-d-glucose (in which the hydrogen atom on C-2 was replaced by a fluorine atom) met these requirements. An electrophilic reaction with ¹⁸F using elemental fluorine ([¹⁸F]F₂) produced via the ²⁰Ne(d,α)¹⁸F reaction on Brookhaven’s 152-cm (60-in) cyclotron (4) was developed and used to produce ¹⁸F-FDG in sufficient yield for transport from Long Island to the University of Pennsylvania for the first human studies in 1976 (1,5).

Over the next 10 years, the demand for ¹⁸F-FDG grew. This created the need for a simpler and higher-yield radiosynthesis that would be more amenable to automation and regional distribution. Indeed, the electrophilic radiosynthesis had shortcomings.

[¹⁸F]F₂ production was technically complex and required the addition of highly toxic and reactive fluorine gas. With [¹⁸F]F₂, only a maximum 50% of the ¹⁸F produced by the ²⁰Ne(d,α)¹⁸F reaction could be incorporated into the final product; this synthesis also gave a mixture of isomers, reducing the overall yield. Fortunately, ¹⁸F (as fluoride ion) could also be produced in far higher yields via the ¹⁸O(p,n)¹⁸F reaction than via the ²⁰Ne(d,α)¹⁸F reaction. Production of ¹⁸F also did not require the addition of fluorine gas (4). This set the stage for intense competition between different groups of chemists to produce ¹⁸F-FDG via a nucleophilic displacement reaction with [¹⁸F]fluoride ion. Several different nucleophilic routes were explored from 1976 to 1986 (4). However, similar to the electrophilic route, all were plagued with difficult steps, including low yields in the incorporation of ¹⁸F and difficulty in removal of protective groups.

A major advance in the synthesis of ¹⁸F-FDG from [¹⁸F]fluoride was reported in 1986 when Hamacher and colleagues at the Kernforschungsanlage Jülich reported that Kryptofix [2.2.2] (Merck), a phase-transfer catalyst, could be used to increase the reactivity of [¹⁸F]fluoride in nucleophilic displacement reactions (2). The reaction of Kryptofix 222 [¹⁸F]fluoride with 1,3,4,6-tetra-O-acetyl-2-O-trifluoromethanesulfonyl-β-d-manno-pyranose gives 1,3,4,6-tetra-O-acetyl-2-[¹⁸F]fluoro-β-d-gluco-pyranose with a 95% incorporation of ¹⁸F. The overall synthesis, including purification, proceeds in about 60% yield. The synthesis is also technically simple. It involves two steps, displacement of a trifluoromethanesulfonyl group with ¹⁸F with [¹⁸F]fluoride and removal of the acetyl groups with HCl. This synthesis produces a single isomer (as confirmed by ¹⁹F nuclear magnetic resonance of the product from the synthesis with unlabeled fluoride ion) and was an elegant solution to the need to produce ¹⁸F-FDG in high yield and in high purity without added carrier.

Over the past three decades, considerable effort has been put into fine tuning the reaction, developing automated radiosynthesis modules, and identifying impurities and contaminants that are carried through to the final product (4). This need has become more critical with the increasing use of ¹⁸F-FDG in clinical practice, where a pharmaceutical-quality product is required.

In summary, Hamacher’s simple and elegant radiosynthesis has made ¹⁸F-FDG available for distribution from many central production sites around the world for basic and clinical applications in institutions that have a PET (or PET/CT) scanner but no cyclotron or chemistry infrastructure. This radiosynthesis was transformative. ¹⁸F-FDG is now used routinely by many hospitals as an off-the-shelf radiopharmaceutical for clinical research and diagnosis in heart disease, neurologic disorders, and oncology, which is the area of most rapid growth.

• © 2020 by the Society of Nuclear Medicine and Molecular Imaging.

• Received for publication May 24, 2020.
• Accepted for publication May 28, 2020.