BLZ945 derivatives for PET imaging of colony stimulating factor-1 receptors in the brain

Article abstract of DOI:10.1016/j.nucmedbio.2021.06.005

Background: The kinase colony stimulating factor-1 receptor (CSF-1R) has recently been identified as a novel therapeutic target for decreasing tumor associated macrophages and microglia load in cancer treatment. In glioblastoma multiforme (GBM), a high-grade cancer in the brain with extremely poor prognosis, macrophages and microglia can make up to 50% of the total tumor mass. Currently, no non-invasive methods are available for measuring CSF-1R expression in vivo. The aim of this work is to develop a PET tracer for imaging of CSF-1R receptor expression in the brain for future GBM patient selection and treatment monitoring. Methods: BLZ945 and a derivative that potentially allows for fluorine-18 labeling were synthesized and evaluated in vitro to determine their affinity towards CSF-1R. BLZ945 was radiolabeled with carbon-11 by N-methylation of des-methyl-BLZ945 using [¹¹C]CH₃I. Following administration to healthy mice, metabolic stability of [¹¹C]BLZ945 in blood and brain and activity distribution were determined ex vivo. PET scanning was performed at baseline, efflux transporter blocking, and CSF-1R blocking conditions. Finally, [¹¹C]BLZ945 binding was evaluated in vitro by autoradiography on mouse brain sections. Results: BLZ945 was the most potent compound in our series with an IC₅₀ value of 6.9 ± 1.4 nM. BLZ945 was radiolabeled with carbon-11 in 20.7 ± 1.1% decay corrected radiochemical yield in a 60 min synthesis procedure with a radiochemical purity of >95% and a molar activity of 153 ± 34 GBq·μmol^?1. Ex vivo biodistribution showed moderate brain uptake and slow wash-out, in addition to slow blood clearance. The stability of BLZ945 in blood plasma and brain was >99% at 60 min post injection. PET scanning demonstrated BLZ945 to be a substrate for efflux transporters. High brain uptake was observed, which was shown to be mostly non-specific. In accordance, in vitro autoradiography on brain sections revealed high non-specific binding. Conclusions: [¹¹C]BLZ945, a CSF-1R PET tracer, was synthesized in high yield and purity. The tracer has high potency for the target, however, future studies are warranted to address non-specific binding and tracer efflux before BLZ945 or derivatives could be translated into humans for brain imaging.

Full text of DOI:10.1016/j.nucmedbio.2021.06.005

Nuclear Medicine and Biology 100–101 (2021) 44–51
Contents lists available at ScienceDirect
Nuclear Medicine and Biology
journal homepage: www.elsevier.com/locate/nucmedbio
BLZ945 derivatives for PET imaging of colony stimulating factor-1
☆
receptors in the brain
Berend van der Wildt a,b, Zheng Miao a, Samantha T. Reyes a, Jun H. Park a, Jessica L. Klockow a, Ning Zhao a,
Alex Romero a, Scarlett G. Guo a, Bin Shen a, Albert D. Windhorst b, Frederick T. Chin a,
⁎
a
Molecular Imaging Program at Stanford (MIPS), Department of Radiology, Stanford University, School of Medicine, Stanford, CA, USA
b
Amsterdam UMC, Vrije Universiteit Amsterdam, Radiology & Nuclear Medicine, de Boelelaan 1117, Amsterdam, Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history:
Background: The kinase colony stimulating factor-1 receptor (CSF-1R) has recently been identified as a novel ther-
apeutic target for decreasing tumor associated macrophages and microglia load in cancer treatment. In glioblas-
toma multiforme (GBM), a high-grade cancer in the brain with extremely poor prognosis, macrophages and
microglia can make up to 50% of the total tumor mass. Currently, no non-invasive methods are available for mea-
suring CSF-1R expression in vivo. The aim of this work is to develop a PET tracer for imaging of CSF-1R receptor
expression in the brain for future GBM patient selection and treatment monitoring.
Methods: BLZ945 and a derivative that potentially allows for fluorine-18 labeling were synthesized and evaluated
in vitro to determine their affinity towards CSF-1R. BLZ945 was radiolabeled with carbon-11 by N-methylation of
des-methyl-BLZ945 using [11C]CH3I. Following administration to healthy mice, metabolic stability of [11C]BLZ945
in blood and brain and activity distribution were determined ex vivo. PET scanning was performed at baseline, ef-
flux transporter blocking, and CSF-1R blocking conditions. Finally, [11C]BLZ945 binding was evaluated in vitro by
autoradiography on mouse brain sections.
Received 16 March 2021
Received in revised form 4 May 2021
Accepted 17 June 2021
Keywords:
Positron emission tomography
Colony stimulating Factor-1 receptor
Tumor associated macrophages
Glioblastoma
Companion diagnostic
BLZ945
Results: BLZ945 was the most potent compound in our series with an IC50 value of 6.9 1.4 nM. BLZ945 was
radiolabeled with carbon-11 in 20.7 1.1% decay corrected radiochemical yield in a 60 min synthesis procedure
with a radiochemical purity of >95% and a molar activity of 153
34 GBq·μmol−1. Ex vivo biodistribution
showed moderate brain uptake and slow wash-out, in addition to slow blood clearance. The stability of BLZ945
in blood plasma and brain was >99% at 60 min post injection. PET scanning demonstrated BLZ945 to be a sub-
strate for efflux transporters. High brain uptake was observed, which was shown to be mostly non-specific. In ac-
cordance, in vitro autoradiography on brain sections revealed high non-specific binding.
Conclusions: [11C]BLZ945, a CSF-1R PET tracer, was synthesized in high yield and purity. The tracer has high po-
tency for the target, however, future studies are warranted to address non-specific binding and tracer efflux be-
fore BLZ945 or derivatives could be translated into humans for brain imaging.
© 2021 Published by Elsevier Inc.
1. Introduction
disease.
A new strategic target for GBM treatment is the colony stimulating
Glioblastoma multiforme (GBM) is a grade IV brain cancer with ex-
tremely poor prognosis; chemotherapy alone or combined with radia-
tion therapy and surgery only prolongs the median survival time of
GBM patients from 10 months to 14 months [1–3]. Newer therapies,
such as bevacizumab, approved as a second line treatment for patients
with recurrent GBM, and local drug-release implants introduced after
surgical resection, only provide limited effects [4–6]. Therefore, novel
treatment strategies are essential for combatting this devastating
factor 1 receptor (CSF-1R), which is expressed on tumor-associated mi-
croglia and macrophages (TAMs). Together TAMs can comprise up to
50% of the tumor mass [7,8]. Microglia and macrophages depend on
the activation of CSF-1R by their endogenous ligand Colony Stimulating
Factor-1 (CSF-1) for their survival and proliferation. In the absence of
CSF-1 or the presence of CSF-1R inhibitors macrophage and microglia
are unable to differentiate and will eventually be depleted from the tis-
sue. This strategy of CSF-1R inhibition has resulted in reduced GBM
tumor mass and a corresponding prolonged survival in GBM mouse
models and thus provides a promising new approach to combatting
brain cancers [9]. Currently, many small molecule and antibody inhibi-
tors of CSF-1R are in (pre)clinical development [10,11]. In addition to
cancer, CSF-1R has been implicated as a pharmacological target for
neuroinflammatory and neurodegenerative diseases [12].
☆
Given his role as Editor in Chief of this journal, Albert D. Windhorst had no involve-
ment in the peer-review of this article.
⁎
Corresponding author at: 3165 Porter Drive, CA, 94304 Palo Alto, USA.
E-mail address: chinf@stanford.edu (F.T. Chin).
https://doi.org/10.1016/j.nucmedbio.2021.06.005
0969-8051/© 2021 Published by Elsevier Inc.

B. van der Wildt, Z. Miao, S.T. Reyes et al.
Nuclear Medicine and Biology 100–101 (2021) 44–51
Positron emission tomography (PET) is a molecular imaging tech-
nique that allows for the quantitative and non-invasive assessment of
biological processes such as transporter and enzyme activity and recep-
tor density [13]. A validated CSF-1R selective PET tracer would have
enormous value in facilitating study of brain diseases, selecting appro-
priate patient therapies, monitoring treatment responses, and develop-
ing new drugs. Three CSF-1R PET tracers have been reported to date
(Fig. 1) [14–16]. However, these tracers suffer from various limitations,
which could prevent their clinical translation. E.g. [18F]10 has poor affin-
ity and low selectivity [14], whereas [11C]AZD6495 displayed insuffi-
cient brain uptake [15]. The most promising CSF-1R tracer reported to
date is [11C]CPPC, which has a high affinity for CSF-1R and displays
good brain penetration in mice and baboon [16]. However, this com-
pound is derived from a rather non-selective compound [17] and in ad-
dition, shows high non-specific binding in the brain.
Among the multiple CSF-1R small molecule inhibitors currently in
clinical development [17–22] is BLZ945, a highly potent and selective
CSF-1R inhibitor (Ki = 1 nM, > 3200 fold selectivity over other kinases)
[9]. This CSF-1R inhibitor demonstrated great efficacy in GBM mice, im-
plicating good brain uptake. Taken together, BLZ945 has high potential
for translation into the first highly potent and selective CSF-1R PET
tracer, either by direct labeling with carbon-11 or by development of an-
alogues that allow for fluorine-18 labeling. The aim of the current study
is to develop a new CSF-1R PET tracer based on BLZ945 as a companion
diagnostic for imaging CSF-1R expression in the brain.
to the following scheme (Method A): 0 min, 35% B; 2 min, 35% B; 20
min, 60% B. UV detection was performed using a Knauer k-2001 UV de-
tector. Radioactivity levels were determined using a dose calibrator
(Capintec CRC-15 PET). Analytical HPLC was performed on an Agilent
Technologies 1200 HPLC system (Agilent Technologies, Santa Clara,
CA, USA) connected to a Raytest GABI* radiodetector (Elisia Raytest
GmbH, Straubenhardt, Germany, sensitivity 370 Bq (10 nCi)) using a
Luna 5u C18 column (250 × 4.6 mm, 5 μm particle size) using a mixture
of A) H2O and B) MeCN as a mobile phase according to the following
scheme (Method B): 0 min, 50% B; 15 min, 70% B; 15.1 min, 95% B; 17
min, 95% B; 17.1 min, 50% B; 17.1–20.0 min 50% B. Mice were housed
in 12 h light/dark cycles and were provided food and water ad libitum.
Blood was obtained by arterial puncture and centrifuged using an ultra-
centrifuge system (Eppendorf Minispin Plus). Gamma counting was
performed on a Hidex Automatic Gamma Counter (Turku, Finland).
TLC plates were exposed to a phosphorimager screen (Fujifilm, Valhalla,
NY, USA) for 30 min in a cassette (Hypercasette, Amersham Biosciences,
Little Chalfont, United Kingdom) and subsequently developed using a
Typhoon Trio Imager (GE Healthcare, Waukesha, WI, USA). Images
were analyzed using ImageJ version 1.49v (National Institute of Health,
USA). PET scanning was performed on a Siemens Inveon dPET and Sie-
mens Inveon Hybrid MicroPET/CT (Munich, Germany) with identical
PET components. Scans were acquired in list mode and rebinned into
the following frame sequence: 7.5 s, 4 × 15 s, 37.5 s, 3 × 60 s, 180 s, 9
× 300 s and 435 s. Following a transmission scan for attenuating correc-
tion, reconstruction was performed (3D OSEM). Images were analyzed
using PMOD (PMOD Technologies LLC, Zurich, Switzerland) by drawing
regions of interest over selected organs and tissues. PLX3397 was pur-
chased from Selleckchem. CPPC and ‘compound 22’ were synthesized
using reported literature procedures [16,23]. Spectroscopic and spectro-
metric data (1H NMR, 13C NMR and ESI-HRMS) was in accordance with
reported values (data not shown).
2. Material and methods
2.1. General
All chemical reagents and solvents were obtained from commercial
suppliers and used as received. Microwave reactions were performed
on a CEM Discover Legacy (Matthews, NC, USA). Reaction monitoring
was performed by thin layer chromatography on pre-coated silica 60
F254 aluminum plates (Merck, Darmstadt, Germany). Spots were visu-
alized with UV light (254 nm), KMnO4, or ninhydrin staining. NMR spec-
troscopy was performed using an Agilent 400 MR (Agilent Technologies,
Santa Clara, CA, USA) with chemical shifts (δ) reported in parts per mil-
lion (ppm) relative to the solvent (CDCl3 1H: 7.26 ppm, 13C: 77.16 ppm;
DMSO‑d6 1H: 2.50 ppm, 13C 39.52 ppm; CD3OD 1H: 3.31 ppm, 13C: 49.00
ppm). High resolution mass spectrometry was performed on a Bruker
Daltonics - apex-Qe (Bruker, Billerica, MA, USA) in either positive or
negative ion mode. Enzyme inhibition assay kits (Z-LYTE™ assay) and
human recombinant enzymes were obtained from ThermoFisher Scien-
tific (Waltham, MA, USA). Fluorescent readout of 384-well plates was
performed on a fluorimeter (Safire, Tecan, XFluo). Carbon-11 was pre-
pared by the 14N(p,α)11C nuclear reaction on a GE PETtrace 880 cyclo-
tron and was delivered as [11C]CO2 using nitrogen as a carrier gas to
the experimental set up (GE TRACERlab FX-C Pro, Boston, MA, USA).
Preparative HPLC was performed on a Dionex P680 HPLC pump
equipped with a Phenomenex Gemini semi preparative HPLC column
(250 × 10 mm, 5u) using a mixture of A) H2O and B) MeCN according
2.2. Chemistry
Compound 1 ((1R,2R)-2-((6-methoxybenzo[d]thiazol-2-yl)amino)
cyclohexan-1-ol):
A solution of 2-chloro-6-methoxybenzo[d]thiazole (1.0 g, 5.0 mmol),
(1R,2R)-2-aminocyclohexan-1-ol (0.76 g, 5.0 mmol), TEA (1.7 mL, 12.5
mmol) in NMP (2 mL) was heated at 120 °C for 48 h. The reaction mix-
ture was cooled to rt. and the precipitates were removed by filtration.
The filtrate was concentrated in vacuo and purified by flash column
chromatography (EtOAc/hexanes 1:1) to afford the product as a white
solid (0.82 g, 59%). 1H NMR (400 MHz, CDCl3): δ = 7.39 (d, 1H, J =
9.1 Hz), 7.05 (d, 1H, J = 2.5 Hz), 6.86 (dd, 1H, J = 2.6, 8.8 Hz), 5.81
(bs, 1H), 3.79 (s, 3H), 3.44 (m, 2H), 2.11 (m, 2H), 1.70 (m, 2H),
1.41–1.16 (m, 4H), 0.27 (s, 1H). 13C NMR (101 MHz, CDCl3): δ =
166.81, 155.35, 145.93, 131.43, 119.21, 113.58, 105.47, 75.28, 61.81,
55.99, 34.27, 32.07, 24.86, 24.15. ESI-HRMS: m/z calculated for
C
14H18N2O2S: 278.1089; found: 301.0980 [M + Na]+
Compound 2 (2-(((1S,2R)-2-hydroxycyclohexyl)amino)benzo[d]
thiazol-6-ol).
.
Fig. 1. Previously reported CSF-1R PET tracers [14–16] and BLZ945 as a potential PET tracer with improved characteristics (this work). The positions of the carbon-11 labels are depicted
with *.
45

B. van der Wildt, Z. Miao, S.T. Reyes et al.
Nuclear Medicine and Biology 100–101 (2021) 44–51
To a solution of 1 (0.56 g, 2.0 mmol) and TEA (0.56 mL, 4.0 mmol) in
DCM (5 mL) at 0 °C was added boron tribromide (1.0 M in DCM, 4.0
mL, 4.0 mmol) and the resulting solution was stirred for 3 h while
allowing gradual warming to room temperature. After concentration
in vacuo the residue was purified by flash column chromatography
(EtOAc/hexanes 2:1) to afford the product as a white solid (0.48 g,
90%). 1H NMR (400 MHz, CD3OD): δ = 7.22 (d, 1H, J = 8.7 Hz), 6.99
(d, 1H, J = 2.1 Hz), 6.73 (dd, 1H, J = 2.2, 8.6 Hz), 3.52 (m, 1H), 3.42
(m, 1H), 2.17 (m, 1H), 2.01 (m, 1H), 1.73 (m, 2H), 1.41–1.21 (m, 4H).
13C NMR (101 MHz, CD3OD): δ = 167.59, 153.70, 146.27, 132.03,
119.09, 114.91, 107.85, 74.50, 61.49, 35.38, 32.59, 25.60, 25.27. ESI-
HRMS: m/z calculated for C13H16N2O2S: 264.0932; found: 265.1006 [M
132.83, 120.08, 119.70, 115.19, 114.69, 110.77, 83.08 (d, J = 167.7 Hz),
74.35, 61.66, 41.06 (d, J = 21.3 Hz), 35.44, 32.53, 25.62, 25.29.
4-((2-(((1S,2R)-2-hydroxycyclohexyl)amino)benzo[d]thiazol-6-yl)
oxy)-N-methylpicolinamide, BLZ945:
To
a solution of 3 (25 mg, 63 μmol) in THF was added
methylamine·HCl (21 mg, 0.30 mmol) and K2CO3 (43 mg, 0.30
mmol). The mixture was reacted for 16 h at 50 °C. After filtration and
concentration in vacuo, the product was purified by flash column chro-
matography (EA - > 2% MeOH in EA) to obtain the product as an off-
white solid (20 mg, 80%). 1H NMR (400 MHz, CDCl3): δ = 8.35 (d, 1H,
J = 5.6 Hz), 8.02 (m, 1H), 7.68 (d, 1H, J = 2.5 Hz), 7.49 (d, 1H, J = 8.7
Hz), 7.27 (s, 1H), 6.99 (dd, 1H, J = 2.5, 8.7 Hz), 6.92 (dd, 1H, J = 2.6,
5.6 Hz), 6.18 (bs, 1H), 3.72 (bs, 1H), 3.50 (m, 2H), 3.00 (d, 3H, J = 5.2
Hz), 2.11 (m, 2H), 1.74 (m, 2H), 1.44–1.24 (m, 4H). 13C NMR (101
MHz, CDCl3): δ = 168.43, 166.94, 164.75, 152.26, 149.82, 149.77,
148.27, 131.81, 119.65, 119.33, 114.02, 113.52, 110.15, 75.06, 61.88,
34.28, 31.93, 26.29, 24.79, 24.17. ESI-HRMS: m/z calculated for
+ H]+
.
Compound 3 (4-((2-(((1S,2R)-2-hydroxycyclohexyl)amino)benzo
[d]thiazol-6-yl)oxy)picolinamide):
A mixture of 2 (66 mg, 0.25 mmol), methyl 4-fluoropicolinate (39
mg, 0.25 mmol) and K2CO3 (69 mg, 0.50 mmol) in DMF (4 mL) was
heated at 100 °C for 16 h. The mixture was diluted with EtOAc, filtered
and concentrated in vacuo. After flash column chromatography (EA -
> 2% MeOH in EtOAc) the product was obtained as a yellow solid (51
mg, 51%). 1H NMR (400 MHz, CDCl3): δ = 8.54 (d, 1H, J = 5.6 Hz),
7.62 (d, 1H, J = 2.5 Hz), 7.52 (d, 1H, J = 8.6 Hz), 7.28 (d, 1H, J = 2.4
Hz), 7.00 (dd, 1H, J = 2.5, 8.7 Hz), 6.97 (dd, 1H, J = 2.5, 5.6 Hz), 5.9
(m, 2H), 3.96 (s, 3H), 3.51 (m, 2H), 2.18 (m, 2H), 1.74 (m, 2H), 1.33
(m, 4H). 13C NMR (101 MHz, CDCl3): δ = 168.37, 166.55, 165.53,
151.41, 149.97, 148.21, 131.94, 119.87, 119.37, 114.73, 113.51, 113.39,
113.38, 75.24, 61.88, 53.13, 34.35, 32.02, 24.81, 24.15. ESI-HRMS: m/z
C
20H22N4O3S: 398.1413; found: 421.1308 [M + Na]+
.
2.3. In vitro IC50 determination
Affinities for designed compounds were determined using a com-
mercially available Z'-LYTE™ assay according to the manufacturer's in-
structions. Briefly, to a 384 well plate were added the respective
kinase (either CSF-1R, PDGFR-β or c-KIT in a final concentration of 0.5,
1.0 and 1.0 ng/μL, respectively) in supplied kinase buffer, ATP (in a
final concentration of 50 μM for CSF-1R and PDGFR-β, 100 μM for c-
KIT), the respective inhibitor (2-fold dilution series in a concentration
range from 1 μM to 0.5 nM), and the assay FRET-peptide. The resulting
solutions (total volume of 10 μL per well) were incubated at room
temperature for 1 h, followed by the addition of protease solution (5
μL per well). After careful mixing, the solutions were incubated for 1 h
followed by the addition of stop buffer (5 μL per well). Readout of the
plate was performed using excitation wavelength of 400 nm and
emission wavelengths of 445 nm (coumarin) and 520 nm (fluorescein)
with a 12 nm bandwidth. Negative control and positive control reac-
tions were performed by omission of ATP and inhibitor, respectively.
The obtained values are the result of three independent experiments
and are expressed as average IC50 value standard deviation.
calculated for C20H21N3O4S: 399.1253; found: 422.1145 [M + Na]+
.
Compound 4 4-((2-(((1S,2R)-2-hydroxycyclohexyl)amino)benzo
[d]thiazol-6-yl)oxy)picolinamide.
A mixture of 2 (66 mg, 0.25 mmol), 4-fluoropicolinamide (35 mg,
0.25 mmol) and K2CO3 (69 mg, 0.50 mmol) in DMF (4 mL) was heated
at 100 °C for 16 h. The mixture was diluted with EtOAc, filtered and con-
centrated in vacuo. After flash column chromatography (EA - > 2%
MeOH in EA) the product was obtained as a yellow solid (80 mg, 80%).
1H NMR (400 MHz, CDCl3): δ = 8.38 (d, 1H, J = 5.6 Hz), 7.86 (d, 1H, J
= 4.2 Hz), 7.67 (d, 1H, J = 2.6 Hz), 7.48 (d, 1H, J = 8.7 Hz), 7.27 (m,
1H), 6.97 (m, 2H), 6.24 (bs, 1H), 6.07 (d, 1H, J = 4.1 Hz), 4.32 (bs,
2H), 3.49 (m, 2H), 2.17 (m, 2H), 1.73 (m, 2H), 1.43–1.23 (m, 4H). 13C
NMR (101 MHz, CDCl3): δ = 168.42, 166.92, 166.62, 151.81, 150.02,
149.98, 148.17, 131.86, 119.67, 119.32, 114.43, 113.52, 110.41, 74.93,
61.90, 34.28, 31.92, 24.79, 24.19. ESI-HRMS: m/z calculated for
2.4. Radiosynthesis
C
19H20N4O3S: 384.1256; found: 385.1331 [M + H]+
Compound N-(2-fluoroethyl)-4-((2-(((1S,2R)-2-hydroxycy-
.
After proton bombardment, [11C]CO2 was delivered to a TRACERlab
FXC-pro module and trapped on a mixture of molecular sieves, pre-
charged with hydrogen gas, and nickel. After completing delivery, the
molecular sieve trap was heated to 350 °C and the formed [11C]CH4
was released to a silica trap, cooled to −80 °C using liquid nitrogen.
This trap was heated to release [11C]CH4 into a closed circuit, where it
was mixed with iodine vapor and converted to [11C]CH3I at 720 °C.
The [11C]CH3I that was formed was accumulated on a cooled Porapak
trap. When the radioactivity levels on this Porapak trap reached a pla-
teau, it was heated to 220 °C to release [11C]CH3I, which was transferred
to a reaction vial using a gentle stream of helium (20 mL/min). The reac-
tion vial was previously charged with precursor 4 (1.0 mg, 2.6 μmol) and
TBAOH (10 μL, 1.0 M in MeOH) in DMSO (400 μL) and was loaded with
5
clohexyl)amino)benzo[d]thiazol-6-yl)oxy)picolinamide.
A solution of compound 3 (40 mg, 0.10 mmol) in MeOH/THF/KOH (4
M in H2O) (1 mL, 4:9:2, v/v/v) was stirred at room temperature for 1 h.
The solution was concentrated in vacuo, diluted with water (10 mL) and
then acidified to pH 3 with 1 M HCl. The mixture was extracted with
DCM (3 × 10 mL). The combined organic fractions were concentrated
to dryness and the residue was resuspended in DMF (2 mL). Then,
DiPEA (39 mg, 0.30 mmol), 2-fluoroethylamine·HCl (10 mg, 0.10
mmol) and HATU (45 mg, 0.10 mmol) were added and the solution
was left for 16 h at room temperature. After concentration in vacuo,
the residue was purified by flash column chromatography (2% MeOH
in DCM) to afford the product as a white solid (25 mg, 58%). Intermedi-
ate: 1H NMR (400 MHz, CD3OD): δ = 8.62 (d, 1H, J = 6.1 Hz), 7.75 (m,
2H), 7.64 (d, 1H, J = 8.7 Hz), 7.39 (m, 2H), 3.55 (m, 2H), 2.14 (m, 2H),
1.82 (m, 2H), 1.43 (m, 4H); 13C NMR (101 MHz, CD3OD): δ = 171.25,
170.23, 163.84, 150.59, 149.69, 147.97, 140.89, 127.96, 121.88, 117.38,
116.50, 116.42, 114.62, 74.55, 63.97, 35.40, 31.83, 25.51, 25.10. Title
compound 5: 1H NMR (400 MHz, CD3OD): 8.46 (d, 1H, J = 5.6 Hz),
7.54 (s, 1H), 7.47 (d, 1H, J = 8.6 Hz), 7.41 (s, 1H), 7.04 (m, 2H), 4.61
(t, 1H, J = 5.1 Hz), 4.49 (t, 1H, J = 5.1 Hz), 3.66 (m, 3H), 3.46 (m, 1H),
2.20 (m, 1H), 2.06 (m, 1H), 1.78 (m, 2H), 1.37 (m, 4H); 13C NMR (101
MHz, CD3OD): 169.77, 168.46, 166.52, 153.01, 151.54, 151.45, 149.31,
[
11C]CH3I until radioactivity reached a maximum. The reactor was
sealed and the reaction solution was heated at 100 °C for 4 min and
subsequently cooled to 20 °C. Next, the reaction solution was diluted
with H20 (1.0 mL) and purified using HPLC using method A. The
collected HPLC product fraction was diluted with H2O (40 mL) and
passed over a preconditioned solid phase extraction cartridge (C18
Sep-Pak Light, Waters). After washing the cartridge with H2O (10 mL),
the product was obtained by sequential elution with ethanol (1.0 mL)
and saline (9.0 mL). A sample was analyzed by analytical HPLC to deter-
mine the (radio)chemical purity and molar activity (Method B).
46

B. van der Wildt, Z. Miao, S.T. Reyes et al.
Nuclear Medicine and Biology 100–101 (2021) 44–51
2.5. Partition coefficient LogD
anesthesia (2% in O2 at 1 L/min). At the indicated time-points blood
was collected by arterial puncture, approximately 0.5 mL) and trans-
ferred to a Heparin coated Eppendorf tube. Mice were perfused gently
(25 mL of phosphate buffer) to remove the blood component from or-
gans. Organs were collected, weighed, and counted using a gamma
counter. Results are expressed as percent of injected dose per gram (%
ID/g) standard deviation.
The partitioning of [11C]BLZ945 between 1-octanol and 0.2 M phos-
phate buffer (pH = 7.4) was determined by vigorously mixing [11C]
BLZ945 (100 μL, 20 MBq) with a solution of 0.2 M phosphate buffer (2
mL, pH 7.4) and 1-octanol (2 mL) for 1 min using a vortex apparatus.
After a settling period of 1 h, three samples of 100 μL were taken from
both layers. Samples were counted for radioactivity and the Log D values
were calculated according to the following formula: Log Doct,7.4 = Log
(Aoct/Aphosphate buffer), where Aoct and Aphosphate buffer represent average
radioactivities of three 1-octanol and three phosphate buffer samples,
2.9. PET and PET/CT scanning
Mice were anesthetized with 2% isofluorane gas in O2 at 1 L/min and
catheterized in their tail vein. When indicated, cyclosporin A was ad-
ministered by tail vein injection 30 min prior to the start of PET scanning
at 30 mg/kg, formulated at 20 mg/mL in DMSO, EtOH, polyethylene gly-
col 400, and H2O (1/15/50/34 v/v/v/v). Unlabeled BLZ945 was formu-
lated in 10% EtOH in saline and administered by co-injection with
respectively. The result is expressed as mean
(n = 3).
standard deviation
2.6. Plasma stability
[
11C]BLZ945 (50 μL, 10 MBq) was added to freshly thawed (500 μL)
[
11C]BLZ945. PLX3397 (Selleckchem) was dosed at 5 mg/kg, formulated
mouse plasma (Sigma Aldrich, catalogue number P9275) and incubated
at 37 °C for 1 h. Ice-cold acetonitrile (1.0 mL) was added and the mixture
was centrifuged for 5 min at 15,000 RPM. An aliquot of the supernatant
was analyzed by analytical HPLC using method B (retention time of [11C]
BLZ945 is 8.4 min).
by dissolving in the cyclosporin A solution and administered by tail vein
injection 30 min prior to tracer administration. [11C]BLZ945 was admin-
istered (10–15 MBq in 100 μL) by tail vein injection. After PET scanning,
mice were sacrificed by cervical dislocation and brains were collected,
weighed, and activity levels were determined using a gamma counter.
Time-activity curves were obtained by drawing regions of interest
using PMOD. Results are expressed as the average %ID/g standard de-
viation (n = 4 per experimental group).
2.7. Radiometabolite analysis
Mice (10–16 weeks old, 25–30 g, n = 3 per time point) were injected
with [11C]BLZ945 (100 μL, approximately 10 MBq) under isoflurane an-
esthesia (2% in O2 at 1 L/min). At the indicated time-points, blood was
collected by arterial puncture, approximately 0.5 mL) and transferred
to a Heparin coated Eppendorf tube. Mice were perfused gently (25
mL of phosphate buffer) and brains (left hemisphere) were collected.
Blood samples were centrifuged for 5 min at 4600 RPM to separate
blood plasma from cells. The plasma (100 μL) was diluted in acetonitrile
(200 μL at 0 °C) and centrifuged for 5 min at 15,000 RPM for removal of
proteins. Brains were homogenized in acetonitrile (200 μL at 0 °C) and
centrifuged for 5 min at 15,000 RPM. An aliquot of each supernatant
(10 μL) was transferred to a TLC plate, which was subsequently dried
at room temperature for 5 min and ran in a solution of DCM/MeOH/
TEA (90:10:1, v/v/v). The radioTLC plate was transferred to a
phosphorimager storage screen and left for 1 h. Readout was performed
on a Typhoon phosphorimager and subsequent analysis was performed
using ImageJ.
2.10. Autoradiography
Autoradiography was performed in flash frozen mouse brain sec-
tions (10 μm thickness). Sections were washed with 50 mM Tris-HCl
buffer (pH 7.4) for 15 min. After drying under a gentle air flow the sec-
tions were incubated with [11C]BLZ945 (0.5 MBq·mL−1) in 50 mM Tris-
HCl, pH 7.4 with 3% BSA in the absence or presence of a CSF-1R inhibitor
at 1 μM concentration for 30 min. Washing was performed with cold
Tris-HCl (5 mM, 4 °C, two times) followed by dipping in ice cold
water. After drying in an air stream, mouse brain sections were exposed
to a phosphorimaging screen (GE Healthcare, Buckinghamshire, UK) for
30 min and developed on a Typhoon FLA 7000 phosphor imager (GE
Healthcare, Buckinghamshire, UK). Visualisation of binding was per-
formed using ImageQuantTL v8.1.0.0 (GE Healthcare, Buckinghamshire,
UK). Data are expressed as % of binding relative to [11C]BLZ945 binding
in the absence of CSF-1R inhibitor (n = 3).
2.8. Biodistribution of [11C]BLZ945
2.11. Statistical analysis
Mice (10–16 weeks old, 25–30 g, n = 3 per time point) were injected
with [11C]BLZ945 (100 μL, approximately 10 MBq) under isoflurane
Statistical analysis was performed using a one-sided, unpaired Stu-
dent's t-test.
Scheme 1. Synthesis of BLZ945, analogue 5 and precursor 4 for carbon-11 radiolabeling. Reagents and conditions: a) (1R,2R)-2-aminocyclohexan-1-ol, TEA, NMP, 120 °C, 48 h, 59%; b) BBr3,
TEA, DCM, 0 °C to rt., 3 h, 90%; c) methyl 4-fluoropicolinate, K2CO3, DMF, 100 °C, 16 h, 51%; d) methylamine·HCl, K2CO3, THF, 50 °C, 16 h, 80%; e) 4-fluoropicolinamide, K2CO3, DMF, 100 °C,
16 h, 80%; f) i) KOH, MeOH/THF/H2O, rt., 1 h; ii) 2-fluoroethylamine·HCl, HATU, DiPEA, rt., 16 h, 58%.
47

B. van der Wildt, Z. Miao, S.T. Reyes et al.
Nuclear Medicine and Biology 100–101 (2021) 44–51
at this position. In accordance with the general selectivity for this scaffold,
no inhibition of c-KIT and PDGFR-β was observed at concentration up to 4
μM [27]. As a result of this affinity and selectivity screening, BLZ945 was
selected as the optimal candidate for radiolabeling.
3.3. Radiolabeling, QC and in vitro evaluation
Carbon-11 synthesis of BLZ945 was pursued by direct methylation of
precursor amide 4. Although this precursor has an aminocyclohexanol
functionality that might benefit from protection to avoid the formation
of side products, such strategy would require multi-step syntheses and
the resulting elongated synthesis time might not outweigh the improved
radiochemical conversion to [11C]BLZ945. Following a one-step strategy,
[
[
11C]BLZ945 was efficiently synthesized by reacting precursor 4 with
11C]CH3I in the presence of tetrabutylammonium hydroxide (1.0 M solu-
tion in MeOH) in DMSO (Fig. 3). The main radiochemical product was
[
11C]BLZ945, the identity of which was confirmed by coinjection of an au-
Fig. 2. Inhibition of compounds towards CSF-1R.
thentic reference sample on HPLC using two distinct HPLC conditions
(Supporting Information Fig. S1). The full synthetic procedure, from end
of bombardment to formulation in 10% EtOH in saline, was performed
in a fully automated fashion in 60 min and gave [11C]BLZ945 in 20.7
3. Results and discussion
1.1% decay corrected yield (2.58
molar activity of 153
34 GBq·μmol−1 and a radiochemical purity
>95% (n = 5, Supporting Information Fig. S1).
11C]BLZ945 remained stable in solution as well as in mouse plasma
up to 1 h (Supporting information Fig. S1). The partition coefficient LogD
was empirically determined as 2.26 0.01 by the shake flask method.
This value is comparable with the calculated LogP of 3.09 (ChemDraw
Professional, v16.0.1.4) and well within the range for successful brain
penetration [29].
0.14 GBq at end of synthesis), a
3.1. Chemistry
Based on the chemical structure, direct radiolabeling of BLZ945 at the
N-methyl functionality with carbon-11 (Fig. 1) is feasible by reacting
[
[
11C]CH3I with the des-methyl precursor molecule [24]. In addition, efforts
were made to design a derivative that allows for fluorine-18 labeling and
thus has improved properties for PET imaging (Scheme 1) [25]. A substitu-
tion of the methyl functionality for a fluoroethyl group was envisioned to
allow for potential fluorine-18 labeling [26]. The synthesis of the target
compounds and their respective precursor molecules for radiolabeling
was based on reported literature procedures [9,27] and is depicted in
Scheme 1. Briefly, 2-chloro-6-methoxybenzothiazole was reacted with
(1R,2R)-2-aminocyclohexanol to obtain benzothiazole 1, which was then
subjected to O-demethylation using BBr3, affording compound 2. Subse-
quent nucleophilic aromatic substitution using 4-fluoropicolinamide re-
sulted in BLZ945 precursor 4. Methyl ester 3 was obtained by reaction of
compound 2 with methyl 4-fluoropiconilate. Compound 3 was then con-
verted to BLZ945 by treatment with methylammonium chloride and to
fluoro-ethyl analogue 5 by alkaline deprotection followed by HATU-
mediated coupling with fluoroethylamine.
3.4. Ex vivo evaluation
Potential metabolism of [11C]BLZ945 was determined ex vivo by
radioTLC analysis of blood plasma and brain homogenates at 15, 30, and
60 min post injection in healthy mice (n = 3 per time point). Both in
plasma and brain only, intact tracer was detected, demonstrating the ex-
cellent metabolic stability of BLZ945 up to 1 h post injection (Supporting
information Fig. S2). These findings correspond well with the reported
high stability of [14C]BLZ945 in human hepatocytes and microsomes
[30]. Ex vivo tracer distribution was determined at 15, 30, and 60 min
post injection in healthy mice (Fig. 4 and Supporting Information
Table S1). In accordance with its pharmaceutical activity in mouse models
of glioblastoma [9], [11C]BLZ945 displayed good brain penetration of ap-
proximately 1%ID/g. Moderate washout was observed, which could po-
tentially be attributed to the basal expression of CSF-1R on microglia in
the healthy brain [31]. Blood radioactivity concentrations were relatively
high, potentially due to the high plasma protein binding (96%) and low
free fraction (4%) of BLZ945 [30]. Excretion mainly occurs via the hepatic
system, as evidenced by high radioactivity concentrations in liver and in-
testines and low values for kidney and urine.
3.2. In vitro evaluation
Both BLZ945 and analogue 5 were tested for their inhibitory potency
in vitro against human recombinant CSF-1R (Fig. 2) using a commercially
available Z-LYTE™ assay kit. In addition, selectivity over c-Kit and PDGFR-
β, close family members of CSF-1R with high structural homology [28],
was determined. BLZ945 was found to be most potent, with IC50 value
of 6.9 1.4 nM. The fluoroethyl analogue 5 resulted in an increase in
IC50 to 40.5 3.7 nM, demonstrating the preference for a methylamide
Fig. 3. Optimization of radiolabeling conditions towards [11C]BLZ945. Reagents and conditions: a) TBAOH (1.0 M in MeOH), DMSO, 100 °C, 5 min.
48

B. van der Wildt, Z. Miao, S.T. Reyes et al.
Nuclear Medicine and Biology 100–101 (2021) 44–51
post injection (p = 0.0003). Specific binding of [11C]BLZ945 was dem-
onstrated by co-administration of unlabeled BLZ945, which resulted in
a 20% decrease in brain activity concentrations at 60 min post injection
(p = 0.05). At time-points up to 8 min post injection, brain radioactivity
levels were higher for the BLZ945 blocking conditions, potentially due
to displacement of [11C]BLZ945 in peripheral organs, resulting in in-
creased blood concentrations. Indeed, higher blood concentrations (de-
termined as heart TACs) were found in this blocking condition at earlier
time points (Supporting information Fig. S3). The suggestion that this
initial increase in brain activity level is due to increased tracer availability
and not a result of CSF-1R binding is further supported by the accelerated
washout of [11C]BLZ945 when compared to unblocked conditions.
PLX3397 pretreatment (i.v. administration, 5 mg/kg, 30 min prior to
tracer injection), did not result in a blocking effect. PLX3397 is designed
to bind at the juxtamembrane domain of CSF-1R in the auto-inhibited
state [32], which is only partially overlapping with the ATP-binding do-
main. The binding domain of BLZ945 has not been reported, but based
on the ATP competitive binding, it is likely that BLZ945 directly occupies
the ATP binding site. The potentially differing binding domains could ex-
plain the lack of a blocking effect of [11C]BLZ945 binding upon PLX3397
pretreatment. Similar lack of blocking has previously been reported
using CSF-1R PET tracer [11C]CPPC. Here, no blocking effect of PLX3397
was observed by in vitro autoradiography studies on human Alzheimer's
disease brain sections [16], whereas BLZ945 co-incubation did result in
anticipated blocking. The brain radioactivity concentrations determined
by PET imaging were confirmed by post-scanning biodistribution
(Fig. 6), which overlapped very well with PET scanning results (Fig. 5B).
The efflux transporter substrate behavior of BLZ945, which is evidenced
by the increased brain uptake upon Cyclosporin A pretreatment, could po-
tentially hamper clinical translation, as efflux transporter inhibition in
human is not recommended.
Fig. 4. Ex vivo tissue distribution of [11C]BLZ945 following administration to healthy mice
(n = 3). Error bars indicate standard deviations (n = 3 per time point).
3.5. In vivo evaluation
Dynamic distribution of [11C]BLZ945 in mice brain was evaluated by
means of PET scanning. Brain uptake was determined in different exper-
imental conditions; besides administration to non-treated mice, [11C]
BLZ945 was administered to mice that were pretreated with cyclospor-
ine A (30 mg/kg, 30 min prior to PET tracer administration). Blocking ex-
periments were performed using excess unlabeled BLZ945 (1 mg·kg−1
,
3.6. Autoradiography
co-injection with radiotracer) and PLX3397 (5 mg/kg, 30 min prior to
PET tracer administration). BLZ945 is a substrate for efflux transporters,
as evidenced by the approximate 2-fold increase of brain activity levels
when comparing baseline with efflux transporter inhibition at 60 min
Finally, in vitro autoradiography on mouse brain sections using [11C]
BLZ945 was performed. Various CSF-1R inhibitors were used for
Fig. 5. In vivo brain uptake of [11C]BLZ945 in healthy mice. A) Time-activity-curves (TACs) of whole mouse brain at different experimental conditions; B) Brain uptake at 60 min post
injection and representative coronal sections of mice from the corresponding experimental groups. Brain regions are indicated with white dotted circles. Results are expressed as
average standard deviation (n = 4).
49

B. van der Wildt, Z. Miao, S.T. Reyes et al.
Nuclear Medicine and Biology 100–101 (2021) 44–51
Fig. 6. Post-mortem brain activity concentrations. Results are expressed as the average
standard deviation (n = 4 per group).
blocking of [11C]BLZ945 binding (Fig. 7) [9,16,23,32].
Fig. 7. In vitro autoradiography with [11C]BLZ945 on mouse brain sections. Representative
brain slices are depicted together with quantification of binding relative to baseline
conditions ([11C]BLZ945 only). CSF-1R inhibitors were used at 1 μM. (* indicates p <
0.05, ** indicates p < 0.005, n.s. not significant, n = 3).
The most pronounced blocking of [11C]BLZ945 binding was ob-
served when using an excess of BLZ945 or when using CPPC as a CSF-
1R inhibitor. The blocking effect was about 30%, indicating specific and
selective binding of [11C]BLZ945 to the brain tissue. However, this result
also demonstrates that about 70% of binding is non-specific. This modest
drop in binding corresponds well with PET imaging results, where a
modest decrease of [11C]BLZ945 binding of approximately 20% was ob-
served. PLX3397 blocking showed a less pronounced blocking effect,
again in close accordance with PET imaging results. Similar results
were obtained using the bisamide CSF-1R inhibitor ‘compound 22’.
Overall, the autoradiography results confirm the high non-specific bind-
ing observed in PET imaging experiments. This non-specific binding was
observed with [11C]CPPC as well and drastically hampers CSF-1R PET
imaging [16]. Given that [11C]BLZ945 is also a strong substrate for efflux
transporters, no further efforts were undertaken to evaluate the com-
pound in animal models of CSF-1R overexpression, such as a glioblas-
toma model. Instead, future work will focus on identifying CSF-1R PET
tracers that do not suffer from efflux transport and less from non-
specific binding compared to the currently reported CSF-1R PET tracers,
which is especially relevant when considering the modest increases in
CSF-1R expression in various animal models [33–36]. To assist in the
compound selection, the use of both in silico (e.g. CNS-MPO-PET scoring)
and in vitro screening tools (e.g. lipid binding assay [37]) are recom-
mended to optimize the tracer development process.
focus on radiolabeling and evaluation of structurally distinct CSF-1R in-
hibitors. Further research on [11C]BLZ945 or fluorine-18 labeled deriva-
tives could focus on imaging of accumulation of macrophages in the
periphery, where efflux transport at the blood-brain-barrier is irrelevant
and non-specific binding in the relevant tissue might be lower (e.g.
cancers, inflammation or atherosclerotic plaques) [38].
Funding
The project was supported, in part, by The Ben and Catherine Ivy
Foundation (FTC), the National Cancer Institute: R21 CA205564 (FTC)
& T32 CA118681 (JLK), and the National Institutes of Health: S10
OD018130 (FTC).
Availability of data and materials
Please contact author for data requests.
Ethics approval and consent to participate
4. Conclusion
Not applicable.
[
11C]BLZ945, a highly potent and selective CSF-1R inhibitor, was suc-
cessfully obtained in high yield and purity. Unfortunately, because of the
strong efflux transporter substrate behavior and high non-specific bind-
ing in the brain, [11C]BLZ945 seems unsuitable as a PET tracer for imag-
ing of CSF-1R expression levels in the brain. Therefore, future work will
Declaration of competing interest
There are no competing interests to report.
50

B. van der Wildt, Z. Miao, S.T. Reyes et al.
Nuclear Medicine and Biology 100–101 (2021) 44–51
[19] Genovese MC, Hsia E, Belkowski SM, Chien C, Masterson T, Thurmond RL, et al. Re-
sults from a phase IIA parallel group study of JNJ-40346527, an oral CSF-1R inhibitor,
in patients with active rheumatoid arthritis despite disease-modifying anti-
rheumatic drug therapy. J Rheumatol. 2015;42:1752–60.
[20] von Tresckow B, Morschhauser F, Ribrag V, Topp MS, Chien C, Seetharam S, et al. An
open-label, multicenter, phase I/II study of JNJ-40346527, a CSF-1R inhibitor, in pa-
tients with relapsed or refractory hodgkin lymphoma. Clin Cancer Res. 2015;21:
1843–50.
Acknowledgments
We thank Tim Doyle and the Stanford Small Animal Imaging Facility
for their technical assistance.
Appendix A. Supplementary data
[21] Scott DA, Bell KJ, Campbell CT, Cook DJ, Dakin LA, Del Valle DJ, et al. 3-amido-4-
anilinoquinolines as CSF-1R kinase inhibitors 2: optimization of the PK profile.
Bioorg Med Chem Lett. 2009;19:701–5.
[22] Leukocyte complexity predicts breast cancer survival and functionally regulates re-
sponse to chemotherapy. Cancer Discov. 2011 Jun;1(1):54–67. https://doi.org/10.
1158/2159-8274.CD-10-0028 Epub 2011 Jun 1.
[23] Ramachandran Sreekanth A, Jadhavar Pradeep S, Miglani Sandeep K, Singh
Manvendra P, Kalane Deepak P, Agarwal Anil K, et al. Design, synthesis and optimi-
zation of bis-amide derivatives as CSF1R inhibitors. Bioorg Med Chem Lett. 2017;15:
2153–60.
[24] Poot AJ, van der Wildt B, Stigter-van Walsum M, Rongen M, Schuit RC, Hendrikse NH,
et al. [11C]Sorafenib: radiosynthesis and preclinical evaluation in tumor-bearing
mice of a new TKI-PET tracer. Nucl Med Biol. 2013;40:488–97.
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.nucmedbio.2021.06.005.
References
[1] Louis DN, Perry A, Reifenberger G, von Deimling A, Figarella-Branger D, Cavenee WK,
et al. The 2016 World Health Organization classification of tumors of the central ner-
vous system: a summary. Acta Neuropathol. 2016;131:803–20.
[2] Johnson DR, O’Neill BP. Glioblastoma survival in the United States before and during
the temozolomide era. J Neurooncol. 2012;107:359–64.
[3] Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJB, et al. Radio-
therapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J
Med. 2005;352:987–96.
[4] Wenger KJ, Wagner M, You SJ, Franz K, Harter PN, Burger MC, et al. Bevacizumab as a
last-line treatment for glioblastoma following failure of radiotherapy, temozolomide
and lomustine. Oncol Lett. 2017;14:1141–6.
[5] Diaz RJ, Ali S, Qadir MG, De La Fuente MI, Ivan ME, Komotar RJ. The role of
bevacizumab in the treatment of glioblastoma. J Neurooncol. 2017;133:455–67.
[6] Chaichana KL, Zaidi H, Pendleton C, McGirt MJ, Grossman R, Weingart JD, et al. The
efficacy of carmustine wafers for older patients with glioblastoma multiforme:
prolonging survival. Neurol Res. 2011;33:759–64.
[25] Dollé F. Carbon-11 and fluorine-18 chemistry devoted to molecular probes for imag-
ing the brain with positron emission tomography. J Labelled Comp Radiopharm.
2013;56:65–7.
[26] Tewson TJ. Synthesis of [18F]Fluoroetanidazole: a potential new tracer for imaging
hypoxia. Nucl Med Biol. 1997;24:755–60.
[27] Novartis AG, 6-O-substituted benzoxazole and benzothiazole compounds and
methods of inhibiting CSF-1R signaling. WO2007121484.
[28] Robinson DR, Wu YM, Lin SF. The protein tyrosine kinase family of the human ge-
nome. Oncogene. 2000;19:5548–57.
[29] Pike VW. PET radiotracers: crossing the blood-brain barrier and surviving metabo-
lism. Trends Pharmacol Sci. 2009;30:431–40.
[7] Badie B, Schartner J. Role of microglia in glioma biology. Microsc Res Tech. 2001;54:
[30] Krauser JA, Jin Y, Walles M, Pfaar U, Sutton J, Wiesmann M, et al. Phenotypic and
metabolic investigation of a CSF-1R kinase receptor inhibitor (BLZ945) and its phar-
macologically active metabolite. Xenobiotica. 2015;45:107–23.
[31] Akiyama H, Nishimura T, Kondo H, Ikeda K, Hayashi Y, McGeer PL. Expression of the
receptor for macrophage colony stimulating factor by brain microglia and its upreg-
ulation in brains of patients with Alzheimer’s disease and amyotrophic lateral sclero-
sis. Brain Res. 1994;639:171–4.
[32] Tap William D, Wainberg Zev A, Anthony Stephen P, Ibrahim Prabha N, Zhang Chao,
Healey John H, et al. Structure-guided blockade of CSF1R kinase in Tenosynovial
Giant-cell tumor. N Engl J Med. 2015;373:428–37.
[33] Wang Y, Berezovska O, Fedoroff S. Expression of colony stimulating factor-1 receptor
(CSF-1R) by CNS neurons in mice. J Neurosci Res. 1999;57:616–32.
[34] Murphy GM, Zhao F, Yang L, Cordell B. Expression of macrophage colony stimulating
factor receptor is increased in the AbetaPP(V717F) transgenic mouse model of
Alzheimer’s disease. Am J Pathol. 2000;157:890–5.
[35] Raivich G, Haas S, Werner A, Klein MA, Kloss C, Kreutzberg GW. Regulation of MCSF
receptors on microglia in the normal and injured mouse central nervous system: a
quantitative immunofluorescence study using confocal laser microscopy. J Comp
Neurol. 1998;395:342–58.
[36] Luo Jian, Elwood Fiona, Britschgi Markus, Villeda Saul, Zhang Hui, Ding Zhaoqing,
et al. Colony-stimulating factor 1 receptor (CSF1R) signaling in injured neurons facil-
itates protection and survival. J Exp Med. 2013;210:157–72.
106–13.
[8] Li R, Li H, Yan W, Yang P, Bao Z, Zhang C, et al. Genetic and clinical characteristics of
primary and secondary glioblastoma is associated with differential molecular sub-
type distribution. Oncotarget. 2015;6:7318–24.
[9] Pyonteck SM, Akkari L, Schuhmacher AJ, Bowman RL, Sevenich L, Quail DF, et al. CSF-
1R inhibition alters macrophage polarization and blocks glioma progression. Nat
Med. 2013;19:1264–72.
[10] Cannarile MA, Weisser M, Jacob W, Jegg AM, Ries CH, Rüttinger D. Colony-
stimulating factor 1 receptor (CSF1R) inhibitors in cancer therapy. J Immunother
Cancer. 2017;5:53.
[11] Peyraud F, Cousin S, Italiano A. CSF-1R inhibitor development: current clinical status.
Curr Oncol Rep. 2017;19:70.
[12] Walker DG, Tang TM, Lue LF. Studies on colony stimulating factor receptor-1 and li-
gands colony stimulating factor-1 and interleukin-34 in Alzheimer’s disease brains
and human microglia. Front Aging Neurosci. 2017;9:244.
[13] Ametamey SM, Honer M, Schubiger PA. Molecular imaging with PET. Chem Rev.
2008;108:1501–16.
[14] Bernard-Gauthier V, Schirrmacher R. 5-(4-((4-[(18)F]Fluorobenzyl)oxy)-3-
methoxybenzyl)pyrimidine-2,4-diamine: a selective dual inhibitor for potential
PET imaging of Trk/CSF-1R. Bioorg Med Chem Lett. 2014;24:4784–90.
[15] Tanzey SS, Shao X, Stauff J, Arteaga J, Sherman P, Scott PJH, et al. Synthesis and initial
in vivo evaluation of [11C]AZ683 - a novel PET radiotracer for colony stimulating fac-
tor 1 receptor (CSF1R). Pharmaceuticals. 2018;11:136.
[37] Assmus F, Seelig A, Gobbi L, Borroni E, Glaentzlin P, Fischer H. Label-free assay for the
assessment of nonspecific binding of positronemission tomography tracer candi-
dates. Eur J Pharm Sci. 2015;79:27–35.
[16] Horti AG, Naik R, Foss CA, Minn I, V Misheneva Y Du, Wang Y, et al. PET imaging of
microglia by targeting macrophage colony-stimulating factor 1 receptor (CSF1R).
PNAS. 2018;116:1686–91.
[17] Illig CR, Chen J, Wall MJ, Wilson KJ, Ballentine SK, Jonathan M, et al. Discovery of
novel FMS kinase inhibitors as anti-inflammatory agents. Bioorg Med Chem Lett.
2008;18:1642–8.
[38] Zhao N, van der Wildt B, Shen B, Chin FT. 18F-BLZ945 derivative with macrophage
avidity facilitates PET imaging in inflamed atherosclerotic plaques. WMIC meeting
abstract. Cardiovasc Imag from Basic Transl. 2019 (September 5).
[18] Conway JG, McDonald B, Parham J, Keith B, Rusnak DW, Shaw E, et al. Inhibition of
colony-stimulating-factor-1 signaling in vivo with the orally bioavailable cFMS ki-
nase inhibitor GW2580. Proc Natl Acad Sci U S A. 2005;102:16078–83.
51