Development of highly affine fluorinated ligands and 18F-labelled radiotracers for PET imaging of the adenosine A2A receptor

Adenosine is an essential neuromodulatory molecule that acts via four G-protein coupled receptors (A1R, A2AR, A2BR, A3R). In the central nervous system (CNS), the A2AR is highly concentrated in the striatum. The A2AR is a promising target for positron emission tomography (PET) imaging of neurodegenerative diseases, such as Huntington’s disease (HD), Alzheimer’s disease (AD) and Parkinson’s disease (PD). Istradefylline is the first A2AR antagonist that is approved by the U.S. Food and Drug Administration (FDA) for adjunctive treatment in patients with PD. So far, [18F]MNI-444 [Ki (hA2AR) = 2.8 nM] is the only 18F-labelled A2AR radiotracer evaluated in healthy subjects.

Aiming at the development of A2AR radiotracers with improved molecular imaging properties, this study is based on three recently published lead compounds with a pyrazolo[3,4-d]pyridine, a morpholinobenzo[d]thiazol-2-amine and a pyrazolo[4,3‑e]‑1,2,4-triazolo[1,5‑ c]-pyrimidine scaffold. Herein, a series of 30 fluorinated derivatives was developed by systematic modification of selected lead compounds. The binding affinities torwards the A2AR and the adenosine A1 receptor (A1R) subtypes were determined by in vitro radioligand binding assays. Regarding the binding affinity and selectivity, PYP1 [Ki (hA2AR) = 5.29 nM, Ki (hA1R) = 220 nM)], PYP2 [Ki (hA2AR) = 2.13 nM, Ki (hA1R) = 147 nM)] und TOZ1 [Ki (hA2AR) = 1.00 nM, Ki (hA1R) = 618 nM)] were radiolabelled as the most suitable A2AR ligands in order to perform first preclinical studies in mice. Additionally, FLUDA [Ki (hA2AR) = 0.61 nM, Ki (hA1R) = 767 nM] was developed and radiolabelled based on the known PET radiotracer [18F]FESCH. [18F]FESCH was selected as reference compound and thus, its radio-synthesis was established as well as optimised in our laboratories.

Three different labelling strategies have been investigated in the frame of this work: (i) two-step one-pot radiolabelling procedures using 18F-labelled prosthetic groups, (ii) alcohol-enhanced copper-mediated one-step radiolabelling procedures starting from boronic acid pinacol ester precursors and (iii) conventional one-step radiolabelling procedures starting from nitro precursors. After the successful radiosynthesis, all five A2AR radiotracers were evaluated by in vitro and in vivo experiments. In vitro autoradiography on mice brain slices revealed specific binding of [18F]PPY2, [18F]TOZ1 and [18F]FLUDA in the region of interest (striatum). Metabolism studies in mice showed a fast metabolic degradation of [18F]PPY1 and [18F]PPY2 with the formation of brain penetrating radiometabolites. In contrast, [18F]TOZ1 and [18F]FLUDA displayed a higher metabolic stability in vivo as the reference [18F]FESCH. PET studies of [18F]PPY1, [18F]PPY1 and [18F]TOZ1 in CD-1 mice revealed no specific accumulation in striatum which would be non-consistent with the known A2AR distribution pattern. The findings indicate that these radiotracers may not demonstrate sufficient affinity in vivo for PET imaging of the A2AR in the brain. In contrast to the previous results, striatum was clearly visualized in PET studies with [18F]FLUDA. Altogether [18F]FLUDA revealed improved molecular imaging properties compared [18F]FESCH which might be a result of the introduction of deuterium atoms in the [18F]fluoroethyl chain, thus resulting in an increased metabolic stability.

In conclusion, the preclinical evaluation of the new developed radiotracers demonstrated that [18F]FLUDA has the highest potential to provide information about the A2AR expression by PET imaging of the brain. Hence, we focus on the clinical translation of [18F]FLUDA to study the A2AR expression in patients with Parkinson’s disease.