Improved in vivo PET imaging of the adenosine A2A receptor in the brain using [18F]FLUDA, a deuterated radiotracer with high metabolic stability

Article abstract of DOI:10.1007/s00259-020-05164-4

Purpose: The adenosine A_2A receptor has emerged as a therapeutic target for multiple diseases, and thus the non-invasive imaging of the expression or occupancy of the A_2A receptor has potential to contribute to diagnosis and drug development. We aimed at the development of a metabolically stable A_2A receptor radiotracer and report herein the preclinical evaluation of [¹⁸F]FLUDA, a deuterated isotopologue of [¹⁸F]FESCH. Methods: [¹⁸F]FLUDA was synthesized by a two-step one-pot approach and evaluated in vitro by autoradiographic studies as well as in vivo by metabolism and dynamic PET/MRI studies in mice and piglets under baseline and blocking conditions. A single-dose toxicity study was performed in rats. Results: [¹⁸F]FLUDA was obtained with a radiochemical yield of 19% and molar activities of 72–180 GBq/μmol. Autoradiography proved A_2A receptor–specific accumulation of [¹⁸F]FLUDA in the striatum of a mouse and pig brain. In vivo evaluation in mice revealed improved stability of [¹⁸F]FLUDA compared to that of [¹⁸F]FESCH, resulting in the absence of brain-penetrant radiometabolites. Furthermore, the radiometabolites detected in piglets are expected to have a low tendency for brain penetration. PET/MRI studies confirmed high specific binding of [¹⁸F]FLUDA towards striatal A_2A receptor with a maximum specific-to-non-specific binding ratio in mice of 8.3. The toxicity study revealed no adverse effects of FLUDA up to 30 μg/kg, ~ 4000-fold the dose applied in human PET studies using [¹⁸F]FLUDA. Conclusions: The new radiotracer [¹⁸F]FLUDA is suitable to detect the availability of the A_2A receptor in the brain with high target specificity. It is regarded ready for human application.

Full text of DOI:10.1007/s00259-020-05164-4

European Journal of Nuclear Medicine and Molecular Imaging
https://doi.org/10.1007/s00259-020-05164-4
ORIGINAL ARTICLE
Improved in vivo PET imaging of the adenosine A2A receptor
in the brain using [18F]FLUDA, a deuterated radiotracer with high
metabolic stability
Thu Hang Lai1,2 & Magali Toussaint1 & Rodrigo Teodoro1 & Sladjana Dukić-Stefanović1 & Daniel Gündel1
Friedrich-Alexander Ludwig1 & Barbara Wenzel1 & Susann Schröder2 & Bernhard Sattler3 & Rareş-
Petru Moldovan1 & Björn H. Falkenburger4 & Osama Sabri3 & Winnie Deuther-Conrad1 & Peter Brust1
&
Received: 19 August 2020 /Accepted: 13 December 2020
#
The Author(s) 2021
Abstract
Purpose The adenosine A2A receptor has emerged as a therapeutic target for multiple diseases, and thus the non-invasive imaging of
the expression or occupancy of the A2A receptor has potential to contribute to diagnosis and drug development. We aimed at the
development of a metabolically stable A2A receptor radiotracer and report herein the preclinical evaluation of [18F]FLUDA, a
deuterated isotopologue of [18F]FESCH.
Methods [18F]FLUDA was synthesized by a two-step one-pot approach and evaluated in vitro by autoradiographic studies as
well as in vivo by metabolism and dynamic PET/MRI studies in mice and piglets under baseline and blocking conditions. A
single-dose toxicity study was performed in rats.
Results [18F]FLUDA was obtained with a radiochemical yield of 19% and molar activities of 72–180 GBq/μmol.
Autoradiography proved A2A receptor–specific accumulation of [18F]FLUDA in the striatum of a mouse and pig brain.
In vivo evaluation in mice revealed improved stability of [18F]FLUDA compared to that of [18F]FESCH, resulting in the absence
of brain-penetrant radiometabolites. Furthermore, the radiometabolites detected in piglets are expected to have a low tendency for
brain penetration. PET/MRI studies confirmed high specific binding of [18F]FLUDA towards striatal A2A receptor with a
maximum specific-to-non-specific binding ratio in mice of 8.3. The toxicity study revealed no adverse effects of FLUDA up
to 30 μg/kg, ~ 4000-fold the dose applied in human PET studies using [18F]FLUDA.
Conclusions The new radiotracer [18F]FLUDA is suitable to detect the availability of the A2A receptor in the brain with high
target specificity. It is regarded ready for human application.
.
.
.
Keywords Adenosine receptors A2 A receptor
Thu Hang Lai, Magali Toussaint, Winnie Deuther-Conrad and Peter
Brust contributed equally to this work.
.
Neurodegeneration Positron-emission tomography
.
Fluorine-18 FESCH
This article is part of the Topical Collection on Preclinical Imaging
* Thu Hang Lai
Introduction
t.lai@hzdr.de
* Magali Toussaint
The signaling molecule adenosine is an important modulator of
neurotransmission, regulating physiological processes such as
sleep, motor activity, or sensorimotor gating [1]. Consequently,
modulating adenosine signaling is an emerging treatment op-
tion for neuropsychiatric and neurodegenerative disorders [1].
Adenosine regulates neurotransmission of glutamate, acetyl-
choline, γ-aminobutyric acid, and dopamine using four G-
protein-coupled plasma membrane receptors - adenosine A1,
m.toussaint@hzdr.de
1
Department of Neuroradiopharmaceuticals, Institute of
Radiopharmaceutical Cancer Research, Helmholtz-Zentrum
Dresden-Rossendorf, Leipzig, Germany
2
Department of Research and Development, ROTOP Pharmaka Ltd.,
Dresden, Germany
3
Department of Nuclear Medicine, University Hospital Leipzig,
A2A, A2B, and A3 receptors [1]. Whereas A1, A2B and A3
Leipzig, Germany
4
receptors are widely distributed throughout the brain, the A2A
receptor is specifically expressed at high densities in the dorsal
Department of Neurology, Dresden University Medical Center,
Dresden, Germany

Eur J Nucl Med Mol Imaging
and ventral striatum, the main input structures of basal ganglia
circuitry [2]. The binding of adenosine to the A2A receptor
activates protein kinase A (PKA) via G protein–mediated stim-
ulation of adenylyl cyclase and the corresponding increase in
cyclic adenosine monophosphate (cAMP). In addition, PKA-
independent pathways have also been reported [3, 4]. The A2A
receptor forms dimers with itself and with further G-protein-
coupled receptors, in particular D2, mGluR5, CB1, and A1 [3],
which causes changes in A2A density in a wide range of neu-
ropsychiatric and neurodegenerative diseases.
targeted drug candidate vipadenant with [11C]SCH442416
(Fig. 1) were published about 10 years ago [15]. At about the
same time, the first 18F-labeled radiotracers were reported, e.g.,
analogs of SCH442416, such as [18F]MRS5425, (also known
as [18F]FESCH, Fig. 1) [16, 17]. [18F]MNI-444 (Fig. 1) is
another PET radiotracer that was used for imaging of the A2A
receptor in healthy human subjects [18]. To assess the suitabil-
ity of A2A receptor PET for the assessment of changes in the
availability of A2A/D2 heterodimeric receptors in neurodegen-
erative diseases, our group performed dynamic PET studies in a
rotenone-based mouse model of Parkinson’s disease with
[18F]FESCH. However, the study yielded inconclusive data,
at least in part due to the penetration of a non-negligible fraction
of a single radiometabolite into the mouse’ brain [19].
Therefore, we intended to enhance the metabolic stability of
[18F]FESCH by developing a deuterated isotopologue,
FLUDA. To test the hypothesis that a selective substitution
with deuterium improves the imaging properties of the novel
A2A receptor–specific radiotracer [18F]FLUDA (Fig. 1), we
performed a series of preclinical animal studies including
in vivo metabolism and dynamic PET/MRI investigations in
mouse and dynamic PET investigations in piglets to investigate
target specificity and pharmacokinetics in different species. To
further support the transfer of [18F]FLUDA to first-in-human
studies, a preclinical acute toxicity study in rats has been
commissioned as well.
For instance, the measurement of mRNA levels indicated a
reduced A2A receptor density in the caudate and putamen of
patients with Parkinson’s disease (PD) [5], while increased
A2A protein levels were found in PD patients with dyskinesias
[6, 7] and in patients with Huntington’s disease (HD) [8].
Given that PD is characterized by a reduced mobility whereas
PD patients with dyskinesias and HD patients show excessive
movements, these findings collectively indicate that A2A is an
important reporter - and modulator - of mobility in human
basal ganglia. Accordingly, the A2A antagonist istradefylline
has been approved for treatment of PD in the USA and in
Japan [3, 9].
Molecular imaging of the A2A receptor by means of
positron-emission tomography (PET) has the potential to quan-
titatively assess the receptor availability and changes thereof
during the course of neuropsychiatric diseases’ pathological
processes and determine optimal dosing regiments for drugs
targeting A2A. Accordingly, research teams in both academia
and industry have been working on the development of suitable
PET radiotracers since more than 15 years. Initially, 11C-
labeleld caffeine derivatives such as [11C]KF17837 [10],
[11C]CSC [11], or [11C]KF21213 [12] were developed and
investigated in animal models. With [11C]SCH442416, one
of the first non-xanthine A2A receptor antagonists has been
proven as suitable for in vivo imaging [13]. The first studies
in humans assessing the distribution of A2A receptors in normal
human brain with [11C]KF18446 (also known as [11C]TMSX,
Fig. 1) [14] as well as the occupancy of the A2A receptor of the
Materials and methods
The full description of all procedures is provided in the
Supplementary information (SI).
Chemical synthesis FLUDA was synthesized by a
microwave-assisted alkylation reaction of 4-[3-(5-amino-2-fu-
ran-2-yl-pyrazolo[4,3-e][1,2,4]triazolo[1,5-c]pyrimidin-7-yl)-
propyl]-phenol (desmethyl SCH442416) and 2-fluoroethyl-
1,1,2,2-d4 4-methylbenzenesulfonate (3), that was previously
O
11CH3
11CH3
N
O
N
N
18F
N
N
O
O
N
N
N
O
N
N
N
O
O
N
N
N
O
N
N
O
N
N
NH2
11
18
11
TMSX
MNI-444
[ F]
[
C]
SCH442416
C]
[
NH2
D
D
18F
18F
O
O
D
D
N
N
N
N
O
N
N
O
N
N
N
N
N
N
18
18
FESCH
F]
[
FLUDA
[ F]
NH2
NH2
Fig. 1 Representative radiotracers for PET imaging of the A2A receptor and the herein reported [18F]FLUDA

Eur J Nucl Med Mol Imaging
prepared by tosylation of the commercially available ethane-
d4-1,2-diol (1) and fluorination of ethane-1,2-diyl-d4 bis(4-
methylbenzenesulfonate) (2, Scheme 1).
containing adenosine deaminase (ADA), and then with buffer
containing ADA and [18F]FLUDA alone or together with
10 μM ZM241385 to assess non-specific binding of
[18F]FLUDA or together with different concentrations of test
compounds to estimate their binding affinity towards the A2A
receptor. After washing and drying, the slides were exposed to
a phosphor imager plate, and after scanning, the autoradio-
graphic images were analyzed with the AIDA 2.31 software
and inhibition curves created with GraphPad Prism 4.1
(GraphPad Inc., La Jolla, CA).
Radiosynthesis [18F]FLUDA was prepared by a two-step one-
pot manual radiosynthesis of the tosylate precursor 2 and the
phenol precursor desmethyl SCH442416 with [18F]fluoride
in anhydrous acetonitrile (MeCN) in the presence of potassi-
um carbonate and Kryptofix 222 (K222) according to our op-
timized procedure for the radiosynthesis of [18F]FESCH (Fig.
2a) [19].
In vivo metabolism studies In mice, [18F]FLUDA was admin-
istered i.v. as a bolus in the tail vein of awake animals (~ 36
MBq, n = 3). After 15 min, the animals were slightly anesthe-
tized by isoflurane inhalation, and blood samples were taken
by retro-orbital bleeding. The blood plasma was obtained by
centrifugation. The brain was isolated immediately after cer-
vical dislocation and homogenized in water. For analysis by
RP-HPLC, brain homogenates and blood plasma were mixed
with the 4-fold volumes of acetone/water (4/1; v/v), precipi-
tated proteins removed by centrifugation, the supernatants
were concentrated and analyzed by analytical RP-HPLC.
For analysis by micellar-HPLC (MLC), brain homogenates
and blood plasma were mixed with equal volumes of aqueous
sodium dodecyl sulfate and directly injected into the MLC
system.
Physiochemical properties The chemical stability of
[18F]FLUDA was proven in saline, phosphate-buffered saline
(PBS, pH 7.4) and n-octanol by incubation at 37 °C up to
60 min followed by radio-TLC and radio-HPLC analyses.
The logD7.4 value was experimentally determined by the con-
ventional shake-flask method using n-octanol and PBS as par-
tition system (n = 4).
Animals The animal experiments were performed with female
CD-1 mice (10–12 weeks, 26–38 g) obtained from the
Medizinisch-Experimentelles Zentrum (MEZ) at Universität
Leipzig (Leipzig; Germany) and female piglets (6–12 weeks
old, 14–18 kg) obtained from the Lehr-und Versuchsgut of the
Faculty of Veterinary Medicine at Universität Leipzig (LVG;
Oberholz; Germany).
In piglets, [18F]FLUDA was administered i.v. as a bolus in
the auricular vein of the anesthetized animals (~ 203 MBq, n =
2). Arterial blood was sampled over 120 min and the blood
plasma was obtained by centrifugation. For analysis by semi-
preparative RP-HPLC, blood plasma was mixed with the two-
fold volumes of acetone/water (4/1; v/v), precipitated proteins
were removed by centrifugation, and the supernatants were
concentrated and analyzed by semi-preparative RP-HPLC.
In vitro binding assays The test compounds (10 mM stock in
DMSO) were incubated with crude cell membrane homoge-
nates obtained from CHO cells stably transfected with human
A2A receptor or human A1 receptor and with A2A receptor–
specific [3H]ZM241385 or A1 receptor–specific [3H]DPCPX
in an incubation buffer at room temperature (RT). The IC50
values were determined by non-linear regression analysis with
GraphPad Prism 4.1 (GraphPad Inc.; La Jolla; CA), and Ki
values estimated according to the Cheng-Prusoff equation
with KD,ZM241385 = 0.8 nM and KD,DPCPX = 0.45 nM.
Dynamic PET studies in mice PET/MRI scans were performed
using a preclinical PET/MRI system (PET/MRI 1Tesla;
nanoScan®; MEDISO Medical Imaging Systems; Budapest;
Hungary) in CD-1 mice under baseline (n = 4), control (vehi-
cle, n = 8), and blocking (pre-administration of 2.5 mg/kg
tozadenant (also known as SYN-115) or 1.0 mg/kg
In vitro autoradiography Cryosections from mouse and piglet
brain were thawed, dried, and pre-incubated with buffer
Scheme 1 Synthesis of FLUDA, reagents and conditions. a p-TsCl, NEt3, CH2Cl2, 0 °C, 3 h, 68% yield. b TBAF, MeCN/THF, 90 °C, 15 min, 36%
yield. c Desmethyl SCH442416, Cs2CO3, MeOH, microwave heating (1 h, 100 °C, 100 W), 37% yield

Eur J Nucl Med Mol Imaging
istradefylline (also known as KW-6002) i.v. 15 to 8 min be-
fore radiotracer n = 4, respectively) conditions. Dynamic
whole-body animal PET scans were acquired during 60 min
after i.v. administration of [18F]FLUDA (3.1–9.7 MBq, 0.7–
4.5 nmol/kg). T1-weighted imaging was performed after-
wards for anatomical orientation and attenuation correction.
After reconstruction, PET images were analyzed in PMOD
3.9 (PMOD technologies LLC; Zurich; Switzerland), and vol-
umes of interest were applied to the PET series to extract time-
activity curves (TACs). TACs were expressed in standardized
uptake value (SUV).
item FLUDA was administered by a single bolus i.v. injection
at doses of 6 and 30 μg/kg body weight (bw). Mortality,
clinical pathology parameters (hematology and serum chem-
istry), organ weights, and microscopic tissue parameters were
investigated 24 h and 2 weeks after treatment.
Results
Synthesis of the reference compound and affinity investiga-
tion The deuterated derivative of FESCH was obtained by the
use of a deuterated fluoroethyl building block according to the
synthesis described in Scheme 1. Binding studies were per-
formed by competitive radiotracer binding assays and re-
vealed high-affinity binding of FLUDA towards the human
PET studies in piglets PET scans were obtained on a clinical
PET-System (ECAT Exact HR+; Siemens Healthcare GmbH;
Erlangen; Germany) in piglets under control (vehicle, n = 1)
and blocking conditions (2.5 mg tozadenant /kg i.v. 15 min
before radiotracer followed by continuous infusion at 0.9
mg/kg/h for the duration of the study, n = 1). Dynamic PET
scans were acquired during 90 min after i.v. administration of
[18F]FLUDA (178–229 MBq; 0.08–0.16 fmol/kg, for control
and blocking conditions respectively). After reconstruction,
PET images were analyzed in PMOD 3.9 (PMOD technolo-
gies LLC; Zurich; Switzerland).
A2A receptor with a Ki value of 0.74 0.26 nM and negligible
binding towards the human A1 receptor (Ki > 1 μM).
Radiosynthesis, lipophilicity, and chemical stability
[
18F]FLUDA was obtained by a one-pot two-step
radiofluorination as shown in Fig. 2 with a radiochemical
yield of 19 3% (end of bombardment = EOB), a radiochem-
ical purity of ≥ 99%, and molar activities in the range of 72–
180 GBq/μmol (end of synthesis = EOS) within a total syn-
thesis time of 102 4 min (n = 9). The shake-flask method
was used to determine the logD7.4 value of [18F]FLUDA (2.01
0.07). The incubation of [18F]FLUDA in saline, PBS (pH
Toxicity studies in rats The extended single-dose toxicity stud-
ies of FLUDA in male (n = 45) and female (n = 45) outbreed
Wistar rats were performed in the Biological Testing
Laboratory (BTL) in Russia (Study Number 678/19). The test
Fig. 2 a Scheme of the two-step one-pot procedure for the radiosynthesis
of [18F]FLUDA. b Representative chromatograms of isolation of
[
[18F]FLUDA. HPLC chromatograms obtained by co-injection with the
reference compound FLUDA
18F]FLUDA by semi-preparative HPLC. c Identification of

Eur J Nucl Med Mol Imaging
7.4), and n-octanol has shown no degradation or
characterized with a KD value of 0.68 nM and a Bmax value
defluorination.
of 218 fmol/mg wet weight (Fig. 5c).
In vivo metabolism At 15 min after i.v. injection of
[18F]FLUDA in mice and analysis, the parent fraction
accounted for about 100% (RP-HPLC, recovery: 98%) and
95% (MLC) in brain (Fig. S2); and 71% (RP-HPLC, recovery:
86%, Fig. 3) and 56% (MLC, Fig. S1) in plasma. In piglets,
the parent fraction accounted for about 47% in plasma at
16 min p.i (recovery: 89%, Fig. S1). The metabolic pattern
in mice contained two fractions of radiometabolites, [18F]M1
and [18F]M2, not able to cross the blood-brain barrier (BBB).
In contrast to the in vivo stability of [18F]FLUDA in mice, the
plasma samples of piglets contained two additional
radiometabolites, [18F]M3 and [18F]M4, supposed to have a
similar structure to [18F]FLUDA based on their chromato-
graphic behavior as shown in Fig. 3. Due to the PET studies
in piglets (Fig. 7), these two radiometabolites are not expected
to cross the BBB.
Dynamic PET studies in mice Representative PET images of
[
18F]FLUDA in a mouse brain are shown in Fig. 6a.
Corresponding to the in vitro autoradiography, the highest
uptake was detected in the A2A receptor–rich striatum. Also,
the negligible accumulation of activity in A2A receptor–poor
regions, such as cerebellum, is in agreement with the in vitro
data. These general findings are confirmed by the analysis of
the PET-derived regional time-activity curves (TACs) pre-
sented in Fig. 6b. In fact, a high specific-to-non-specific bind-
ing ratio was reached, with a maximum SUVrSt/Cb of 8.3
reflecting a high target selectivity (Fig. 6c). Furthermore, the
continuous washout of activity from all brain regions confirms
the absence of brain-penetrant radiometabolites. To validate
the target specificity of [18F]FLUDA in vivo, the A2A receptor
antagonist istradefylline (1 mg/kg) or tozadenant (2.5 mg/kg)
were administered i.v. at 10 min before radiotracer injection.
Maximum blocking was obtained by pre-treatment of
istradefylline as reflected by the striatal SUV being similar
to the cerebellar SUV under blocking conditions during the
entire period of investigation (Fig. 6d–e). By tozadenant pre-
treatment, the accumulation of activity in the striatum was
reduced by 23.5% in comparison to the control (according to
the area under the curve (AUC); AUCvehicle: 24.9 8.7 SUV ∙
min; AUCtozadenant: 19.0 4.3 SUV ∙ min from 0 to 60 min;
Fig. S8). The activity accumulation in the cerebellum was not
significantly affected by the blocking compound (Fig. 6e;
AUCvehicle: 5.9 3.0 SUV ∙ min; AUCtozadenant: 5.4 1.3
SUV ∙ min from 0 to 60 min; Fig. S8), confirming the target
specificity of [18F]FLUDA and indicating the suitability of
the cerebellum as a reference region for A2A receptor imaging.
Further analysis of the whole-body PET data obtained in
control (vehicle-treated) animals revealed a high initial uptake
in the small intestine and the liver, followed by a pronounced
accumulation of activity in the former, along with a stable
uptake in the kidney over the whole scanning time, indicating
Quantitative in vitro autoradiography The binding pattern of
[18F]FLUDA in mouse brain is heterogeneous with the
highest density of binding sites in the striatum, an A2A
receptor–rich region. In A2A receptor–poor regions, such as
cerebellum, midbrain, cortex, or thalamus, only negligible
binding was detected (Fig. 4a, Fig. S3). About 90% of the
binding of ~ 1 nM [18F]FLUDA in mouse striatum could be
displaced by co-incubation with 10 μM ZM241385 (Fig. 4b).
The binding sites in striatum were further characterized by
saturation experiments which revealed a KD value of 4.30
0.73 nM and a Bmax value of 556 143 fmol/mg wet weight
(Fig. 4d). Comparable results were obtained in a single exper-
iment performed in cryosections of the pig brain to promote
the preclinical evaluation of [18F]FLUDA in larger species
(Fig. 5). The nearly exclusive binding of [18F]FLUDA in
the striatum (Fig. 5a) was completely blocked by co-
administration of ZM241385 (Fig. 5b). By homologous com-
petition, the binding of FLUDA in the pig striatum was
Fig. 3 Representative RP-HPLC
radiochromatograms of plasma
samples after administration of
[18F]FLUDA to a mouse and
piglet

Eur J Nucl Med Mol Imaging
Fig. 4 Representative in vitro autoradiographic images of the binding
pattern of [18F]FLUDA (0.93 nM) in horizontal mouse brain slices.
The highest accumulation of activity in the striatum (a, red). The
binding is completely blocked by co-administration of 10 μM of the
A2A receptor antagonist ZM241385 (b). For annotation of the brain re-
gions, the slices were Nissl-stained after autoradiography (c). St striatum,
Cb cerebellum, and Cx cortex, Th thalamus. Representative competition
curve (d). KD and Bmax calculated from the homologous completion of
[
18F]FLUDA with FLUDA (Cheng-Prusoff equation KD = IC50
−
[[18F]FLUDA] and Bmax = top – bottom ∙ (KD + [[18F]FLUDA])/
[[18F]FLUDA] [20].
that [18F]FLUDA and/or its metabolites mainly undergo
hepatobiliary excretion (Table S2; Fig. S9).
recommends a minimum of a 50-fold the clinical exposure as
the maximum dose for general toxicity studies in any species,
did not cause any conspicuous features related to the hematol-
ogy, clinical chemistry, necropsy, and histopathology
analyses.
Dynamic PET studies in piglets To further assist the transfer of
[18F]FLUDA to clinical studies, we performed a pilot study to
evaluate the pharmacokinetic profile of [18F]FLUDA by dy-
namic PET in piglets. The activity distribution in the brain of a
control subject resembles the findings obtained in mouse—
strong and nearly exclusive accumulation in the striatum and
low signal in all other brain regions (Fig. 7a). The pre-
administration of tozadenant (2.5 mg/kg) at 15 min before
radiotracer followed by continuous infusion inhibited the ac-
Discussion
The selective deuteration of the known A2A receptor radio-
tracer [18F]FESCH significantly improved the metabolic sta-
bility of the A2A receptor PET ligand [18F]FLUDA.
Furthermore, the preclinical evaluation of [18F]FLUDA in
both small and larger animals demonstrates high specific bind-
ing towards the adenosine A2A receptor in the brain, indicating
the suitability of [18F]FLUDA for first-in-human studies.
In comparison to [18F]FESCH, [18F]FLUDA is much less
susceptible to biotransformation. This is reflected by the much
higher amount of the parent fraction in plasma at 15 min after
i.v. injection in mice (41% [19] vs. 71%). The activity accu-
mulating in the brain after i.v. injection of [18F]FLUDA cor-
responds almost completely to the parent compound. This
cumulation of activity in the striatum by 50% (AUCvehicle
:
42.1 SUV ∙ min; AUCtozadenant: 21.0 SUV ∙ min from 0 to
60 min), and in A2A receptor–poor regions, such as cerebel-
lum by 17% (AUCvehicle: 24.0 SUV ∙ min; AUCtozadenant: 19.8
SUV ∙ min from 0 to 60 min; Fig. 7b).
Single-dose toxicity study In an extended single-dose toxicity
study in rats, administration of FLUDA up to 30 μg/kg body
weight, which is about 4000 times the estimated human dose
and thus in accordance with the ICH guideline M3(R2), which
Fig. 5 In vitro autoradiographic images of the binding pattern of
[
accumulation of activity is in the striatum (a, red, St striatum, Cb
cerebellum, Cx cortex, and Th thalamus). The binding is completely
blocked by co-administration of 10 μM of the A2A receptor antagonist
ZM241385 (b). Homologous competition curve (c)
18F]FLUDA (0.64 nM) in sagittal pig brain slices. The highest

Eur J Nucl Med Mol Imaging
Fig. 6 a Representative
horizontal PET images (0–60
min) of [18F]FLUDA in the brain
of CD-1 mice (striatum: red, cer-
ebellum: yellow). b TACs at
baseline for CD-1 mice (n = 4) in
different brain regions after injec-
tion of [18F]FLUDA; TACs of c
SUVrSt/Cb, d SUV striatum, and e
SUV cerebellum: after pre-
treatment with vehicle (red
square, n = 8), tozadenant (2.5
mg/kg bw, blue circle, n = 4) and
istradefylline (1.0 mg/kg bw,
green triangle, n = 4)
Fig. 7 a Representative horizontal PET images (0–60 min) of [18F]FLUDA in the brain of piglets (striatum: red, cerebellum: yellow). b TACs of SUVrSt/
Cb of [18F]FLUDA in the pig brain after administration of vehicle (red square, n = 1) or blocking with tozadenant (blue circle, n = 1)

Eur J Nucl Med Mol Imaging
finding is supported by the profile of the region-specific TACs
obtained in dynamic PET studies with [18F]FLUDA in both
mice (Fig. 6) and piglets (Fig. 7), which do not indicate any
confounding accumulation of radiometabolites in the brain. In
summary, deuteration of the fluoroethoxy group of
[18F]FLUDA eliminates the main obstacle observed in the
investigation of disease-related changes of the A2A receptor
availability in preclinical studies using [18F]FESCH: signifi-
cant amounts of brain-penetrant radiometabolites.
mg/kg followed by 0.9 mg/kg/h infusion and 2.5 mg/kg, re-
spectively) can be explained by the dose-response results re-
ported for the occupancy of [18F]MNI-444-labeled binding
sites by tozadenant (47% vs. 95% at 1.5 mg/kg and 10.5
mg/kg, respectively; [24]). The observed difference in the
maximum striatum-to-cerebellum ratio between mice (8.3)
and piglet (2.1) is regarded to be caused by the different A2A
receptor densities in the striatum, which were
autoradiographically determined to be twofold higher in mice
than in pigs [25].
With respect to further more general properties such as
affinity, selectivity, blood-brain barrier permeability, and
clearance, [18F]FLUDA retains the positive characteristics of
[18F]FESCH [16, 17, 19, 21]. The affinity of [18/19F]FLUDA
towards human and mouse A2A receptors (Ki ~ 1 nM vs. KD ~
4 nM) corresponds well to the results obtained for [18/
19F]FESCH (Ki ~ 1 nM vs. KD ~ 5 nM; [19]) including a
slight discrepancy between the two species that was reported
previously (Table S1, Figs. S3, S4, S5, S6, S7) [22].
Regarding specificity, the accumulation of [18F]FLUDA re-
sembles the expression pattern of the A2A receptor protein in
both mouse brain and pig brain and both in vitro and in vivo
(Figs. 4, 5, 6, and 7). Moreover, the accumulation of
[18F]FLUDA can be blocked completely by structurally dif-
ferent selective A2A receptor antagonists (Figs. 6 and 7). The
presence of an A2A receptor antagonist inhibited the accumu-
lation of activity in the striatum almost completely but did not
affect any other brain region in mice and affected the cerebel-
lum by 17% in piglet. However, the high specificity of
[18F]FLUDA, as well as the absence of effect of blocking
compound on cerebellum activity accumulation in mice stud-
ies, suggests that this effect on the cerebellum of piglet is most
likely due to inter-individual variability, and would need a
larger-scale study to investigate its significance.
Collectively, these findings indicate highly specific in vivo
binding of [18F]FLUDA in the striata in both species.
The differences in the efficacy of the two A2A receptor
ligands istradefylline and tozadenant to inhibit the binding of
[18F]FLUDA in vitro and in vivo can be explained by known
differences in the properties of these antagonists, in particular
the remarkably low affinity of tozadenant towards mouse A2A
receptors, which we determined in-house by [18F]FLUDA
autoradiography (istradefylline: Ki (mouse A2A receptor) ~
60 nM, Ki (piglet A2A receptor) ~ 15 nM; tozadenant: Ki
(mouse A2A receptor) ~ 250 nM, Ki (piglet A2A receptor) ~
10 nM; Table S1). Most of the preclinical studies that evalu-
ated the specificity of A2A receptor–targeting PET radiotracers
use istradefylline [17, 21, 23], whereas tozadenant has only
been applied, to the best of our knowledge, in a single study in
rhesus monkeys [24]. A head-to-head comparison as present-
ed here for mice has not been reported so far. The higher
efficacy of tozadenant in vivo in piglet in comparison to
mouse (~ 50% displacement vs. ~ 25% displacement at 2.5
By taking into consideration the density of the A2A receptor
in the human striatum under physiologic conditions (e.g., 260–
444 fmol/mg protein [26, 27]), we assume that the binding
potential of [18F]FLUDA is suitable to quantify the receptor
availability in humans. The highly specific binding of
[18F]FLUDA towards the A2A receptor was confirmed by
blocking studies conducted in vitro (Figs. 4 and 5) and
in vivo (Figs. 6 and 7). The presence of A2A receptor antago-
nists inhibited the accumulation of activity in striatum almost
completely but did not affect any other brain region.
Accordingly, from the herein presented exploratory study, we
assume that similar to the already clinically applied non-
deuterated [18F]FESCH or [11C]preladenant, the quantifica-
tion and kinetic modeling of PET data obtained with
[18F]FLUDA might be facilitated by the use of the cerebellum
as a reference region [21, 28]. However, full kinetic modeling
with arterial input functions, ideally performed in a large animal
and under conditions comparable to human studies, is required
to validate the use of such region for the quantification of the
availability of the A2A receptor by [18F]FLUDA PET.
Conclusion
[18F]FLUDA is a new A2A receptor–targeting PET radiotracer
with promising preclinical results for clinical translation. The
radiotracer can be prepared with high molar activities and in
reasonable radiochemical yield (manuscript reporting auto-
mated synthesis under preparation). Highly affine and specific
binding of [18F]FLUDA in vivo was demonstrated by PET
studies performed in different species. The selective deutera-
tion resulted in a high metabolic stability, and the identifica-
tion of cerebellum as a reference region is assumed to facilitate
the quantification of the availability of the A2A receptor in the
brain and disease-related changes thereof in clinical routine.
Furthermore, the results of a toxicity and a dosimetry study
[29] indicate that the use of [18F]FLUDA in first-in-human
imaging studies is safe.
Supplementary Information The online version contains supplementary
material available at https://doi.org/10.1007/s00259-020-05164-4.

Eur J Nucl Med Mol Imaging
Acknowledgments We thank all colleagues of the Institute of Analytical
Chemistry, Department of Chemistry and Mineralogy of Universität
Leipzig (Leipzig; Germany), for NMR and HRMS measurements;
Karsten Franke, Helmholtz-Zentrum Dresden-Rossendorf (HZDR), for
providing fluorine-18; Tina Spalholz, HZDR, for technical assistance;
and Tatjana Sattler and all colleagues of the Department of Veterinary
Medicine and Nuclear Medicine of University Hospital Leipzig (Leipzig;
Germany) for the support of pig studies.
2. Borroto-Escuela DO, Fuxe K. Adenosine heteroreceptor complexes
in the basal ganglia are implicated in Parkinson’s disease and its
treatment. J Neural Transm (Vienna). 2019;126:455–71. https://doi.
org/10.1007/s00702-019-01969-2.
3. Domenici MR, Ferrante A, Martire A, Chiodi V, Pepponi R,
Tebano MT, et al. Adenosine A2A receptor as potential therapeutic
target in neuropsychiatric disorders. Pharmacol Res. 2019;147:
104338. https://doi.org/10.1016/j.phrs.2019.104338.
4. Fredholm BB, Chen JF, Masino SA, Vaugeois JM. Actions of
adenosine at its receptors in the CNS: insights from knockouts
and drugs. Annu Rev Pharmacol Toxicol. 2005;45:385–412.
https://doi.org/10.1146/annurev.pharmtox.45.120403.095731.
5. Hurley MJ, Mash DC, Jenner P. Adenosine A2A receptor mRNA
expression in Parkinson’s disease. Neurosci Lett. 2000;291:54–8.
https://doi.org/10.1016/s0304-3940(00)01371-9.
6. Ramlackhansingh AF, Bose SK, Ahmed I, Turkheimer FE, Pavese
N, Brooks DJ. Adenosine 2A receptor availability in dyskinetic and
nondyskinetic patients with Parkinson disease. Neurology. 2011;76:
1811–6. https://doi.org/10.1212/WNL.0b013e31821ccce4.
Authors’ contributions All authors contributed to the study conception
and design. Material preparation, data collection, and analysis were per-
formed by Thu Hang Lai, Magali Toussaint, Rodrigo Teodoro, Sladjana
Dukić-Stefanović, Friedrich-Alexander Ludwig, Barbara Wenzel, Daniel
Gündel, Bernhard Sattler, Susann Schröder, and Winnie Deuther-Conrad.
The first draft of the manuscript was written by Thu Hang Lai, Magali
Toussaint, Winnie Deuther-Conrad, and Peter Brust, and all authors
commented on previous versions of the manuscript. All authors read
and approved the final manuscript.
7. Villar-Menéndez I, Porta S, Buira SP, Pereira-Veiga T, Díaz-
Sánchez S, Albasanz JL, et al. Increased striatal adenosine A2A
receptor levels is an early event in Parkinson’s disease-related pa-
thology and it is potentially regulated by miR-34b. Neurobiol Dis.
2014;69:206–14. https://doi.org/10.1016/j.nbd.2014.05.030.
8. Varani K, Bachoud-Lévi AC, Mariotti C, Tarditi A, Abbracchio
MP, Gasperi V, et al. Biological abnormalities of peripheral A2A
receptors in a large representation of polyglutamine disorders and
Huntington’s disease stages. Neurobiol Dis. 2007;27:36–43.
https://doi.org/10.1016/j.nbd.2007.03.011.
Funding Open Access funding enabled and organized by Projekt DEAL.
This work (project no. 100226753) was funded by the European Regional
Development Fund (ERDF) and Sächsische Aufbaubank (SAB).
Data availability The datasets generated during and/or analyzed during
the current study are available from the corresponding author on reason-
able request.
Compliance with ethical standards
9. Xu K, Bastia E, Schwarzschild M. Therapeutic potential of adeno-
sine A2A receptor antagonists in Parkinson’s disease. Pharmacol
Ther. 2005;105:267–310. https://doi.org/10.1016/j.pharmthera.
2004.10.007.
Conflict of interest The authors declare that they have no conflicts of
interest.
10. Ishiwata K, Noguchi J, Toyama H, Sakiyama Y, Koike N, Ishii S,
et al. Synthesis and preliminary evaluation of [11C]KF17837, a
selective adenosine A2A antagonist. Appl Radiat Isot. 1996;47:
507–11. https://doi.org/10.1016/0969-8043(95)00295-2.
Ethical approval All animal studies followed the international guide-
lines of animal care and were approved by Landesdirektion Leipzig
(Reg.-Nr.: TVV 18/18; Reference number DD24.1-5131/446/19).
11. Márián T, Boros I, Lengyel Z, Balkay L, Horváth G, Emri M, et al.
Preparation and primary evaluation of [11C]CSC as a possible tracer
for mapping adenosine A2A receptors by PET. Appl Radiat Isot.
1999;50:887–93. https://doi.org/10.1016/s0969-8043(98)00162-6.
12. Wang WF, Ishiwata K, Nonaka H, Ishii S, Kiyosawa M, Shimada J,
et al. Carbon-11-labeled KF21213: a highly selective ligand for
mapping CNS adenosine A2A receptors with positron emission to-
mography. Nucl Med Biol. 2000;27:541–6. https://doi.org/10.
1016/s0969-8051(00)00126-8.
13. Todde S, Moresco RM, Simonelli P, Baraldi PG, Cacciari B,
Spalluto G, et al. Design, radiosynthesis, and biodistribution of a
new potent and selective ligand for in vivo imaging of the adeno-
sine A2A receptor system using positron emission tomography. J
Med Chem. 2000;43:4359–62. https://doi.org/10.1021/
jm0009843.
Consent to participate Not applicable
Consent for publication Not applicable
Code availability Not applicable
Open Access This article is licensed under a Creative Commons
Attribution 4.0 International License, which permits use, sharing,
adaptation, distribution and reproduction in any medium or format, as
long as you give appropriate credit to the original author(s) and the
source, provide a link to the Creative Commons licence, and indicate if
changes were made. The images or other third party material in this article
are included in the article's Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in the
article's Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a
copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
14. Mishina M, Ishiwata K, Kimura Y, Naganawa M, Oda K, Kobayashi
S, et al. Evaluation of distribution of adenosine A2A receptors in
normal human brain measured with [11C]TMSX PET. Synapse.
2007;61:778–84. https://doi.org/10.1002/syn.20423.
15. Brooks DJ, Papapetropoulos S, Vandenhende F, Tomic D, He P,
Coppell A, et al. An open-label, positron emission tomography
study to assess adenosine A2A brain receptor occupancy of
vipadenant (BIIB014) at steady-state levels in healthy male volun-
teers. Clin Neuropharmacol. 2010;33:55–60. https://doi.org/10.
1097/WNF.0b013e3181d137d2.
References
16. Bhattacharjee AK, Lang L, Jacobson O, Shinkre B, Ma Y, Niu G,
et al. Striatal adenosine A2A receptor-mediated positron emission
1. Cunha RA. How does adenosine control neuronal dysfunction and
neurodegeneration? J Neurochem. 2016;139:1019–55. https://doi.
org/10.1111/jnc.13724.

Eur J Nucl Med Mol Imaging
tomographic imaging in 6-hydroxydopamine-lesioned rats using
18F]-MRS5425. Nucl Med Biol. 2011;38:897–906. https://doi.
org/10.1016/j.nucmedbio.2011.01.009.
23. Zhou X, Boellaard R, Ishiwata K, Sakata M, Dierckx R, de Jong JR,
et al. In vivo evaluation of 11C-Preladenant for PET imaging of
adenosine A2A receptors in the conscious monkey. J Nucl Med.
2017;58:762–7. https://doi.org/10.2967/jnumed.116.182410.
24. Barret O, Hannestad J, Alagille D, Vala C, Tavares A, Papin C,
et al. Adenosine 2A receptor occupancy by tozadenant and
preladenant in rhesus monkeys. J Nucl Med. 2014;55:1712–8.
https://doi.org/10.2967/jnumed.114.142067.
25. Sihver W, Schulze A, Wutz W, Stusgen S, Olsson RA, Bier D, et al.
Autoradiographic comparison of in vitro binding characteristics of
various tritiated adenosine A2A receptor ligands in rat, mouse and
pig brain and first ex vivo results. Eur J Pharmacol. 2009;616:107–
14. https://doi.org/10.1016/j.ejphar.2009.06.025.
26. Albasanz JL, Rodríguez A, Ferrer I, Martín M. Adenosine A2A
receptors are up-regulated in Pick’s disease frontal cortex. Brain
Pathol. 2006;16:249–55. https://doi.org/10.1111/j.1750-3639.
2006.00026.x.
27. Wan W, Sutherland GR, Geiger JD. Binding of the adenosine A2
receptor ligand [3H]CGS 21680 to human and rat brain: evidence
for multiple affinity sites. J Neurochem. 1990;55:1763–71. https://
doi.org/10.1111/j.1471-4159.1990.tb04967.x.
[
17. Khanapur S, Paul S, Shah A, Vatakuti S, Koole MJ, Zijlma R, et al.
Development of [18F]-labeled pyrazolo[4,3-e]-1,2,4-triazolo[1,5-
c]pyrimidine (SCH442416) analogs for the imaging of cerebral
adenosine A2A receptors with positron emission tomography. J
Med Chem. 2014;57:6765–80. https://doi.org/10.1021/jm500700y.
18. Barret O, Hannestad J, Vala C, Alagille D, Tavares A, Laruelle M,
et al. Characterization in humans of 18F-MNI-444, a PET radiotrac-
er for brain adenosine 2A receptors. J Nucl Med. 2015;56:586–91.
https://doi.org/10.2967/jnumed.114.152546.
19. Schröder S, Lai TH, Toussaint M, Kranz M, Chovsepian A, Shang
Q, et al. PET imaging of the adenosine A2A receptor in the
rotenone-based mouse model of Parkinson’s disease with
[
18F]FESCH - synthesized by a simplified two-step one-pot
radiolabeling strategy. Molecules. 2020;25. https://doi.org/10.
3390/molecules25071633.
20. Harvey M. The GraphPad guide to analyzing radioligand binding
data. Copyright© 1995-96 by GraphPad Software. http://www3.
uah.es/farmamol/Public/GraphPad/radiolig.htm. Accessed 17
Aug 2020.
21. Khanapur S, Van Waarde A, Dierckx RAJO, Elsinga PH, Koole
MJB. Preclinical evaluation and quantification of 18F-fluoroethyl
and 18F-fluoropropyl analogs of SCH442416 as radioligands for
PET imaging of the adenosine A2A receptor in rat brain. J Nucl
Med. 2017;58:466–72. https://doi.org/10.2967/jnumed.116.
178103.
22. Alnouri MW, Jepards S, Casari A, Schiedel AC, Hinz S, Müller
CE. Selectivity is species-dependent: characterization of standard
agonists and antagonists at human, rat, and mouse adenosine recep-
tors. Purinergic Signal. 2015;11:389–407. https://doi.org/10.1007/
s11302-015-9460-9.
28. Sakata M, Ishibashi K, Imai M, Wagatsuma K, Ishii K, Zhou X,
et al. Initial evaluation of an adenosine A2A receptor ligand, 11C-
preladenant, in healthy human subjects. J Nucl Med. 2017;58:
1464–70. https://doi.org/10.2967/jnumed.116.188474.
29. Sattler B, Kranz M, Lai TH, Gündel D, Toussaint M, Schröder S,
et al. Preclincal incorporation dosimetry of [18F]FLUDA - a novel
18F-labeled tracer for PET imaging of the expression of the adeno-
sine A2A receptor (A2AR). J Nucl Med. 2020;61:1014.
Publisher’s note Springer Nature remains neutral with regard to jurisdic-
tional claims in published maps and institutional affiliations.