18F-Labeled Bombesin Analogs for Targeting
GRP Receptor-Expressing Prostate Cancer
Xianzhong Zhang, PhD1; Weibo Cai, PhD1; Feng Cao, MD, PhD1; Eduard Schreibmann, PhD2; Yun Wu, PhD1;
Joseph C. Wu, MD, PhD1,3; Lei Xing, PhD2; and Xiaoyuan Chen, PhD1
1Molecular Imaging ProgramatStanford(MIPS), Departmentof Radiology andBio-X Program,Stanford University Schoolof Medicine,
Stanford, California;2Department of Radiation Oncology, Stanford University School of Medicine, Stanford, California; and
3Division of Cardiology, Department of Medicine, Stanford University School of Medicine, Stanford, California
The gastrin-releasing peptide receptor (GRPR) is found to be
overexpressed in a variety of human tumors. The aim of this
study was to develop18F-labeled bombesin analogs for PET
of GRPR expression in prostate cancer xenograft models.
Methods: [Lys3]Bombesin ([Lys3]BBN) and aminocaproic acid-
bombesin(7–14) (Aca-BBN(7–14)) were labeled with18F by cou-
pling the Lys3amino group and Aca amino group, respectively,
basic condition (pH 8.5). Receptor-binding affinity of FB-
[Lys3]BBN and FB-Aca-BBN(7–14) was tested in PC-3 human
prostate carcinoma cells. Internalization and efflux of both radio-
tracers were also evaluated. Tumor-targeting efficacy and in vivo
kinetics of both radiotracers were examined in male athymic
nude mice bearing subcutaneous PC-3 tumors by means of
biodistribution and dynamic microPET imaging studies.18F-FB-
[Lys3]BBN was also tested for orthotopic PC-3 tumor delineation.
Metabolicstability of18F-FB-[Lys3]BBNwas determinedinmouse
blood, urine, liver, kidney, and tumor homogenates at 1 h after in-
jection. Results: The typical decay-corrected radiochemical yield
wasabout 30%–40% for both tracers,witha total reactiontimeof
1506 20min starting from18F2.18F-FB-[Lys3]BBNhadmoderate
stability in the blood and PC-3 tumor, whereas it was degraded
rapidly in the liver, kidneys, and urine. Both radiotracers exhibited
rapid blood clearance.18F-FB-[Lys3]BBN had predominant renal
excretion.18F-FB-Aca-BBN(7–14) exhibited both hepatobiliary
and renal clearance.Dynamic microPETimaging studiesrevealed
that the PC-3 tumor uptake of18F-FB-[Lys3]BBN in PC-3 tumor
was much higher than that of18F-FB-Aca-BBN(7–14) at all time
points examined (P , 0.01). The receptor specificity of18F-FB-
[Lys3]BBN in vivo was demonstrated by effective blocking of
tumor uptake in the presence of [Tyr4]BBN. No obvious blockade
was found in PC-3 tumor when18F-FB-Aca-BBN(7–14) was used
as radiotracer under the same condition.18F-FB-[Lys3]BBN was
also able to visualize orthotopic PC-3 tumor at early time points
after tracer administration, at which time minimal urinary bladder
activitywas present to interfere with the receptor-mediated tumor
uptake. Conclusion: This study demonstrates that
tate cancer in vivo.
Key Words: prostate cancer; GRP receptor;
J Nucl Med 2006; 47:492–501
Neuroendocrine (NE) cells are believed to play a para-
crine regulatory role in the prostate (1). Prostatic NE cells
contain abundant secretory granules filled with numerous
bioactive compounds collectively called NE products (NEP)
(2). In particular, members of the gastrin-releasing peptide
(GRP) family and its analog bombesin (BBN) have been
implicated in the biology of several human malignancies,
including lung, colon, breast, and prostate cancers (1–4). To
date, 3 mammalian GRP/BBN receptor subtypes have been
cloned and characterized: the GRP receptor (GRPR), the
BBN-receptor subtype 3 (BRS-3), and the neuromedin-B
receptor (NMBR) (5). Only GRPR was found in prostate
carcinoma (6), although NMBR and BSR-3 have been found
in other cancer types (7,8). Antagonists of GRPR are de-
signed to bind to human GRPR with high affinity and block
the receptor-activated signal transduction pathways and,
thus, inhibit the growth of prostate cancer both invitro and in
vivo (9). GRP/BBN analogs have also been used as carriers
to deliver drugs, radionuclides, and toxins to target prostate
carcinoma and other cancer types that are GRPR positive
(10,11). Therefore, the ability to document GRPR density in
vivo is crucial for the application of GRPR-targeted drug
Being the most widely applied radionuclide for diagnos-
tic purposes, a great deal of research has been done to de-
velop99mTc- and111In-labeled BBN-like peptides involving
a wide range of chelators, peptide sequences, and bifunc-
tional linkers (12). To date, 2 of the de novo radiolabeled
GRP-like peptides, RP527 (13) and the BN1 (14), are under
clinical evaluation with satisfactory results. In addition,90Y,
188Re, and177Lu have been used to radiolabel BBN analogs
for potential peptide receptor radiotherapy applications
PET for cancer imaging of GRPR status in vivo has been
less studied. Rogers et al. developed a truncated form of a
Received Sep. 28, 2005; revision accepted Nov. 15, 2005.
For correspondence or reprints contact: Xiaoyuan Chen, PhD, Molecular
Imaging Program at Stanford (MIPS), Department of Radiology and Bio-X
Program, Stanford University School of Medicine, 1201 Welch Rd., P095,
Stanford, CA 94305-5484.
492THE JOURNAL OF NUCLEAR MEDICINE • Vol. 47 • No. 3 • March 2006
64Cu (half-life [t1/2] 5 12.7 h; b1, 17.4%)-labeled BBN
analog,64Cu-DOTA-Aoc-BBN(7–14) (Aoc is 8-aminooctanoic
acid), for microPETimaging of an androgen-independent PC-3
tumor xenograft model (17). Incorporation of a poly(ethylene
glycol) (PEG) linker (molecular weight 3,400) resulted in
significantly reduced receptor avidity and lower receptor spe-
cific activity accumulation in vivo (18). We recently reported
the synthesis and pharmacologic evaluation of another64Cu-
labeled BBN analog,64Cu-DOTA-[Lys3]BBN, for targeting
GRPR expression in both PC-3 and 22RV1 tumor models
(19). Very recently, another BBN analog, [D-Tyr6,b-Ala11,
Thi13,Nle14]BBN(6–14) amide (BZH3), was conjugated with
(DOTA) through a PEG2 linker and labeled with68Ga (t1/25
68 min; b1, 88%), obtained from a68Ge/68Ga generator for
imaging AR42J rat pancreatic cancer-bearing nude mice (20).
18F (t1/25 109.7 min; b1, 99%) is an ideal short-lived
PET isotope for labeling small molecular recognition units
such as antigen- binding domain of antibody fragments and
small biologically active peptides (21).18F-Labeled pros-
thetic groups such as N-succinimidyl-4-18F-fluorobenzoate
(18F-SFB) have been developed, which can be attached to
either N-terminal or lysine e-amino groups with little or no
loss of bioactivity of the peptide ligand (22,23). In the
present study, we labeled both [Lys3]bombesin ([Lys3]BBN)
and aminocaproic acid-bombesin(7–14) (Aca-BBN(7–14))
with18F for GRPR imaging of subcutaneous and orthotopic
PC-3 tumor models with PET.
MATERIALS AND METHODS
All chemicals obtained commercially were used without further
purification. [Lys3]BBN and Aca-BBN(7–14) were synthesized
using solid-phase Fmoc chemistry by American Peptide, Inc. No-
carrier-added18F2was obtained from PETNET Inc. The received
18F2was trapped on an anion-exchange resin and eluted with 0.5 mL
K2CO3(2 mg/mL in H2O) combined with 1 mL Kryptofix 2.2.2.
(Sigma-Aldrich) (10 mg/mL in acetonitrile). The semipreparative
reversed-phase high-performance liquid chromatography (HPLC)
system has been described elsewhere (24).
Chemistry and Radiochemistry
4-Fluorobenzoyl-bombesin analogs (FB-[Lys3]BBN and FB-
Aca-BBN(7–14)) were synthesized as reference standards. In brief,
an equimolar amount of SFB (in acetonitrile) and [Lys3]BBN or
Aca-BBN(7–14) (in H2O) was mixed and the pH was adjusted
to 8.3 by addition of 0.1N sodium borate buffer. The reaction
mixture was incubated at 40?C for 80 min and then quenched by
trifluoroacetic acid (TFA). Semipreparative HPLC purification
gave the desired products. The HPLC retention times were around
20.8 min for FB-[Lys3]BBN and 19.1 min for FB-Aca-BBN(7–14),
4-18F-Fluorobenzoyl-[Lys3]bombesin (18F-FB-[Lys3]BBN) and
14)) were synthesized by coupling the corresponding BBN peptide
with18F-SFB (25–27).18F-SFB was purified by semipreparative
HPLC, concentrated to about 200 mL, and added to [Lys3]BBN or
Aca-BBN(7–14) peptide (200 mg) in 800 mL of sodium borate
buffer (50 mmol/L, pH 8.5). The reaction mixture was gently
mixed at 40?C for 30 min. Final purification was accomplished
by semipreparative HPLC and the tracers were reconstituted in
phosphate-buffered saline (PBS, pH 7.4) and passed through a
0.22-mm Millipore filter (Millipore Corp.) for in vivo applications.
In Vitro Cell-Binding Assay
The PC-3 human prostate carcinoma cell line was purchased
from American Type Culture Collection. PC-3 cells were grown in
F-12K nutrient mixture (Kaighn’s modification) (Invitrogen Corp.)
supplemented with 10% (v/v) fetal bovine serum (FBS) (Invitrogen
Corp.) at 37?C with 5% CO2. In vitro binding affinity and spec-
ificity of FB-BBN analogs for GRPR were evaluated using com-
petitive receptor-binding assay.
Life Science Products, Inc.; specific activity, 74 TBq/mmol) was
used as the GRPR-specific radioligand. Experiments were performed
as described previously (19). The 50% inhibitory concentration
(IC50) value for the displacement binding of125I-[Tyr4]BBN by
those ligands was calculated by nonlinear regression analysis using
GraphPad Prism software (Graph-Pad Software Inc.). All experi-
ments were performed twice with triplicate samples.
Internalization and Efflux Studies
Internalization and efflux of18F-FB-[Lys3]BBN and18F-FB-
Aca-BBN(7–14) into PC-3 cells were examined following a pro-
tocol reported earlier (19). The data was normalized as percentage
of the total amount of radioactivity added per cell.
All animal experiments were performed under a protocol ap-
proved by the Stanford University Administrative Panel on Lab-
oratory Animal Care (A-PLAC). Both subcutaneous and orthotopic
tumor model were established in 4- to 6-wk-old male athymic
nu/nu mice (Harlan). For the subcutaneous prostate cancer model,
5 · 106PC-3 cells suspended in 50 mL serum-free F-12K medium
and 50 mL Matrigel (BD Biosciences) were injected into the right
shoulder of the mice. For the orthotopic PC-3 tumor model, 5 · 105
cells in 20 mL PBS were injected into the prostate gland of male
nude mice. The prostate of anesthetized mice was exposed through
a midline laparotomy incision and by retraction of the bladder and
male sex accessory glands anteriorly. Injection of cells was per-
formed with a 27-gauge needle inserted into the prostatic lobe
located at the base of the seminal vesicles as described (28). The
abdominal wound was sutured using a 4.0 chromic gut suture in a
The subcutaneous tumor-bearing mice were used for biodis-
tribution when the tumor volume reached 300–400 mm3(3–4 wk
after inoculation). Three mice were each injected intravenously
with about 370 kBq (10 mCi)18F-FB-[Lys3]BBN or18F-FB-Aca-
BBN(7–14). The mice were sacrificed at 60 min after injection
and the body weight was recorded. Blood, tumor, major organs,
and tissues were collected, wet weighed, and counted by
g-counter. The percentage of injected dose per gram (%ID/g) was
determined for each sample. For each mouse, radioactivity of the
tissue samples was calibrated against a known aliquot of radio-
tracer. Values are expressed as mean 6 SD. To test the specific
binding of the radiotracers to PC-3 tumor, GRPR-blocking studies
were performed by examining the biodistribution of each radio-
labeled tracer in the presence of [Try4]BBN as a blocking agent
18F-BOMBESIN PET FOR GRPR • Zhang et al. 493
(10 mg/kg mice body weight). Mice were also sacrificed at 60 min
after injection (n 5 3).
microPET Imaging and Image Analysis
microPET scans were performed on a microPET R4 rodent
model scanner (Concorde Microsystems Inc.). The scanner has a
computer-controlled bed and 10.8-cm transaxial and 8-cm axial
fields of view (FOVs). It has no septa and operates exclusively in
the 3-dimensional (3D) list mode. Animals were placed near the
center of the FOV of the scanner, where the highest image reso-
lution and sensitivity are available. The microPET studies were
performed by tail-vein injection of 3.7 MBq (100 mCi) of
radiotracer (18F-FB-[Lys3]BBN or18F-FB-Aca-BBN(7–14)) under
isoflurane anesthesia. The 60-min dynamic (5 · 1 min, 5 · 2 min,
5 · 3 min, 6 · 5 min) microPET data acquisition (total of 21
frames) was started 4 min after injection. Static images at 2.5-, 3-,
and 4-h time points were also acquired as 10-min static images.
The images were reconstructed by a 2-dimensional ordered-
subsets expectation maximum (OSEM) algorithm and no correc-
tion was necessary for attenuation or scatter (29).
For each microPET scan, regions of interest (ROIs) were drawn
over the tumor, normal tissue, and major organs by using vendor
software (ASI Pro 18.104.22.168) on decay-corrected whole-body coro-
nal images. The maximum radioactivity concentration (accumu-
lation) within a tumor or an organ was obtained from mean pixel
values within the multiple ROI volume, which were converted to
counts/mL/min by using a conversion factor. Assuming a tissue
density of 1 g/mL, the ROIs were converted to counts/g/min and
then divided by the administered activity to obtain an imaging
Male mice bearing PC-3 tumors were injected intravenously
with 3.7 MBq of18F-FB-[Lys3]BBN. The animals were sacrificed
and dissected at 60 min after injection Blood, urine, liver, kidneys,
and tumor were collected. Blood was immediately centrifuged for
5 min at 13,200 rpm. Organs were homogenized using an IKA
Ultra-Turrax T8 (IKAWorks Inc.), suspended in 1 mL of PBS, and
centrifuged for 5 min at 13,200 rpm. After removal of the super-
natants, the pellets were washed with 500 mL of PBS. For each
sample, supernatants of both centrifugation steps were combined
and passed through Sep-Pak C18cartridges. The urine sample was
directly diluted with 1 mL of PBS and passed through Sep-Pak
C18cartridge. The cartridges were washed with 2 mL of H2O
and eluted with 2 mL of acetonitrile containing 0.1% TFA. The
combined aqueous and organic solutions were concentrated to about
1 mL by rotary evaporation, passed through a 0.22-mm Millipore
filter, and injected onto an analytic HPLC column at a flow rate of
1 mL/min using the gradient described earlier. Radioactivity was
monitored using a solid-state radiation detector. At the same time,
the eluent was also collected by a fraction collector (0.5 min/
fraction), and the activity of each fraction was measured by the
g-counter. The HPLC analysis was performed in duplicate and the
extraction efficiency was determined in triplicate. Data obtained
from the g-counter were plotted to reconstruct the HPLC chro-
matograms. Control experiments were also performed to test the
extraction and elution efficiency by adding18F-FB-[Lys3]BBN
directly to the same tissue samples.
To perform a microCT scan, an anesthetized male nude mouse
bearing an orthotopic PC-3 tumor (4–6 wk after inoculation) was
mounted on a turntable bed that could be moved automatically in
the axial direction. The high-resolution 3D images were obtained
by a commercial microCAT II system (ImTek Inc.). This scanner
uses a SourceRay SB-80-50 x-ray tube with about 40-mm focal
spot providing 30-mm resolution in high-resolution configuration.
A total of 220 rotation steps was taken over 220? with one axial
bed position. A standard convolution-backprojection procedure
with a Shepp–Logan filter was used to reconstruct the CT images
in 512 · 512 pixel matrices.
microPET and microCT Image Fusion
For the microPET and microCT coregistration, we used a
narrow-band approach, which is a hybrid method combining the
advantages of pixel-based and distance-based registration tech-
niques (30). In essence, this technique is a 2-step image registra-
tion in which the tumor to be registered is first represented by a
data structure containing the signed distance values from its
boundaries, followed by an image registration using a pixel-based
metric. The optimization aligns the zero set of the narrow band
obtained from the CT images with the tumor gradients in the PET
dataset, eliminating the assumption of uniform pixel intensities
within one organ used in the mutual information (MI) approach. In
our setup, the normalized correlation was used as the metric and a
gradient-based algorithm was used to find the optimal match.
After imaging, both subcutaneous and orthotopic tumors were
dissected for histology to verify tumor pathology. Tumor tissues
were frozen at 280?C in optimal cutting temperature (OCT)
medium. Frozen sections (5 mm; Leica Microsystems, Inc.) were
fixed in acetone at 220?C for 15 min and air-dried overnight
(4?C). They were then stained with hematoxylin–eosin (BD Bio-
sciences). Slides were examined under a ZEISS AxioVert 25 re-
search microscope (Carl Zeiss) equipped with a ZEISS digital
camera (model AxioCam MRc5) and captured with MRGrab
22.214.171.124 (Carl Zeiss vision GmbH) software.
Quantitative data are expressed as mean 6 SD. Means were
compared using 1-way ANOVA and the Student t test. P values ,
0.05 were considered significant.
18F-Fluorination of bombesin analogs ([Lys3]BBN and
Aca-BBN(7–14)) were achieved via
Starting with18F-F2in Kryptofix 2.2.2./K2CO3solution,
the total reaction time, including final HPLC purification,
was about 150 6 20 min. The overall radiochemical yield
with decay correction was 31.4% 6 4.6% (n 5 12). The
radiochemical purity of the labeled peptides was .98% ac-
cording to analytic HPLC. The specific activity of18F-SFB
was estimated by radio-HPLC to be 2002250 TBq/mmol.
18F-FB-[Lys3]BBN and18F-FB-Aca-BBN(7–14) were well
separated from [Lys3]BBN and Aca-BBN(7–14), respec-
tively, rendering the specific activity of these 2 PET tracers
virtually the same as that of18F-SFB.
18F-SFB (Fig. 1).
494THE JOURNAL OF NUCLEAR MEDICINE • Vol. 47 • No. 3 • March 2006
In Vitro Receptor-Binding Assay
The binding affinities of [Lys3]BBN, Aca-BBN(7–14),
FB-[Lys3]BBN, and FB-Aca-BBN(7–14) for GRPR were
evaluated for PC-3 cells. Results of the cell-binding assay
were plotted in sigmoid curves for the displacement of
125I-[Tyr4]BBN from PC-3 cells as a function of increasing
concentration of bombesin analogs. The IC50values were
determined to be 3.3 6 0.4 nmol/L for [Lys3]BBN, 20.8 6
0.3 nmol/L for Aca-BBN(7–14), 5.3 6 0.6 nmol/L for FB-
[Lys3]BBN, and 48.7 6 0.1 nmol/L for FB-Aca-BBN(7–14)
on 1 · 105PC-3 cells. [Lys3]BBN with the full sequence
of the bombesin peptide is substantially more potent than
Aca-BBN(7–14) with the truncated sequence. Coupling of
the fluorobenzoyl group had a minimal effect on the binding
affinity for both compounds.
Internalization and Efflux Studies
The results for the internalization of both tracers,18F-
FB-[Lys3]BBN and18F-FB-Aca-BBN(7–14), are shown in
Figure 2A. For both tracers, internalization occurred during
5 min of incubation after the preincubation step: 51% for
18F-FB-[Lys3]BBN and 58% for18F-FB-Aca-BBN(7–14),
respectively. After approximately 15 min of incubation, in-
ternalization of both tracers reached a maximum (85% for
18F-FB-[Lys3]BBN and 60% for18F-FB-Aca-BBN(7–14))
and then decreased slowly through 120 min of incubation
18F-FB-[Lys3]BBN and 50% for
BBN(7–14) at 120 min). When blocked with 200 mmol/L
of [Tyr4]BBN, the nonspecific uptake for both tracers was
,10% over the incubation period (data not shown).
Efflux studies were performed for up to 3 h of incubation
to further characterize both tracers (Fig. 2B).18F-FB-[Lys3]
BBN and18F-FB-Aca-BBN(7–14) tracers exhibited similar
efflux curves. After 30 min of incubation, approximately
54% of18F-FB-[Lys3]BBN had effluxed out of the cells. At
the end of the 3-h incubation period, approximately 77% of
the radiotracer had effluxed. For18F-FB-Aca-BBN(7–14)
tracer, after 30 min of incubation, approximately 39% of
the radioactivity effluxed out of the PC-3 cells and, after 3 h
of incubation, approximately 83% of the radioactivity had
effluxed. The efflux rate of
faster than that of
18F-FB-[Lys3]BBN, which might be
due to the lower affinity of18F-FB-Aca-BBN(7–14) for the
GRPR than18F-FB-[Lys3]BBN, as determined from the in
vitro cell-binding assay.
Biodistribution of18F-FB-[Lys3]BBN and18F-FB-Aca-
BBN(7–14) was evaluated in athymic nude mice bearing
subcutaneous PC-3 tumors. The results were shown in
Figure 3. For18F-FB-[Lys3]BBN (Fig. 3A), the tumor up-
take was 5.94 6 0.78 %ID/g at 60 min after injection,
which decreased to 0.50 6 0.11 %ID/g in the presence of a
blocking dose of [Tyr4]BBN (10 mg/kg mice body weight).
[Tyr4]BBN was also able to substantially reduce the activity
accumulation in the pancreas, intestines, and kidneys,
demonstrating that these organs are also GRPR positive.
Increased uptake in the lung, liver, and spleen was ob-
served. For18F-FB-Aca-BBN(7–14) (Fig. 3B), the tumor
uptake (0.43 6 0.18 %ID/g at 60 min after injection) was
more than one order of magnitude lower than that for
Schematic structures of
18F-BOMBESIN PET FOR GRPR • Zhang et al. 495
18F-FB-[Lys3]BBN. A blocking dose of [Tyr4]BBN de-
creased the uptake of18F-FB-Aca-BBN(7–14) in the tumor,
pancreas, and intestines, whereas the uptake in the liver,
kidneys, and lung was increased. Tumor-to-nontarget ratios
of18F-FB-[Lys3]BBN were significantly higher than those of
18F-FB-Aca-BBN(7–14) for all organs and tissues examined
(P , 0.001) (Fig. 3C).
Dynamic microPET Imaging of Subcutaneous PC-3
The dynamic microPET scans were performed on the
subcutaneous PC-3 tumor model with18F-FB-[Lys3]BBN
and18F-FB-Aca-BBN(7–14). Selected coronal images at
different time points after administration of the appropriate
tracers are shown in Figure 4 for comparison. Tumor con-
trast was observed as early as 10 min after injection for
both radiotracers. The tumor uptake of18F-FB-[Lys3]BBN
was 3.50, 3.68, and 2.61 %ID/g at 10, 30, and 60 min after
injection, respectively. The tumor-to-contralateral back-
ground (muscle) ratio was 3.95 at 60 min after injection
Time–activity curves derived from the 60-min dynamic
microPET scan showed that18F-FB-[Lys3]BBN was ex-
creted predominantly through the renal route (Fig. 5A).
Liver had low initial uptake (5.15 %ID/g at 5 min after
injection) and the radioactivity was also washed out rapidly
(1.75 %ID/g at 1 h after injection). The activity accumu-
lation in the kidneys was moderately low at early time
points (4.85 %ID/g at 5 min after injection) but rapidly
increased to 47.00 %ID/g at 50 min after injection followed
by a steep decline afterward (28.49 %ID/g at 60 min and
1.01 %ID/g at 2 h after injection). Compared with18F-FB-
18F-FB-Aca-BBN(7–14) had a significantly
lower tumor uptake, which corroborates the biodistribution
results obtained from direct tissue sampling. The tumor
(B) of18F-FB-[Lys3]BBN and18F-FB-Aca-BBN(7–14) using PC-3
cells. Data are from 2 experiments with triplicate samples and
are expressed as mean 6 SD.
Comparison of internalization (A) and efflux rate
FB-Aca-BBN(7–14) (B) in male athymic nude mice bearing
subcutaneous PC-3 tumors. Mice were injected intravenously
with 370 kBq of radioligand with or without the presence
of [Tyr4]BBN at 10 mg/kg mice body weight and euthanized at
60 min after injection. (C) Tumor-to-nontarget ratios of 2 radio-
tracers resulting from the biodistribution are also shown. Data
are presented as mean %ID/g 6 SD (n 5 3).
Biodistribution of18F-FB-[Lys3]BBN (A) and18F-
496THE JOURNAL OF NUCLEAR MEDICINE • Vol. 47 • No. 3 • March 2006
uptake of18F-FB-Aca-BBN(7–14) was 0.92, 0.71, and 0.78
%ID/g at 10, 30, and 60 min after injection, respectively.
Liver had low uptake at all time points (1.35, 3.29, and 1.75
%ID/g at 5, 25, and 60 min after injection, respectively).
The activity accumulation in the kidneys was also low at
early time points (4.77 %ID/g at 5 min after injection) but
increased to 11.19 %ID/g at 45 min after injection and
remained steady over the remaining dynamic scan period.
Figure 4C shows the transaxial microPET images of PC-3
tumor-bearing mice at 1 h after administration of18F-FB-
[Lys3]BBN, with and without coinjection of 10 mg/kg
[Tyr4]BBN. The blocking reduced the tumor uptake to
0.58 %ID/g at 1 h after injection, 4- to 5-fold lower than
that of the control animals.
The metabolic stability of18F-FB-[Lys3]BBN was deter-
enates at 60 min after injection The extraction efficiencies
were 61.4% for the blood, 95.0% for the liver, 91.1% for
the kidneys, and 97.8% for the PC-3 tumor, respectively.
The elution efficiencies of the soluble fractions were 44.4%
for the blood, 39.8% for the liver, 41.5% for the kidneys,
and 95.5% for the PC-3 tumor. HPLC analysis results of
the acetonitrile-eluted fractions are shown in Figure 6. The
average fraction of intact tracer was between 0.7% and
47.2% (Table 1). Incubation of18F-FB-[Lys3]BBN directly
with tissue and organ homogenates revealed that the extrac-
tion efficiency was .90% in all cases, except for the liver,
for which the extraction efficiency was only 67.7%. The
elution efficiency was also .90% for all samples tested.
Although we did not identify the composition of the metab-
olites, we found that all metabolites came off the HPLC
column earlier than those for the parent compound. No
defluoridation of18F-FB-[Lys3]BBN was observed as no visi-
ble bone uptake was observed in any of the microPET scans.
PET and CT Imaging of Orthotopic PC-3 Tumor Model
We also evaluated18F-FB-[Lys3]BBN in orthotopic PC-3
tumor-bearing mice. The representative microPET images
shown in Figure 7A were at 17 min after injection The
orthotopic tumor uptake was calculated to be 2.07 %ID/g
from microPET imaging, which is somewhat lower than
that of subcutaneous PC-3 tumor (3.74 %ID/g at 17 min
after injection). Dynamic scans indicated that the tumor
was clearly visualized between 10 and 30 min, after which
a significant amount of urinary bladder activity interferes
PC-3 tumor on the right shoulder. Coronal images (decay
corrected to time of tracer injection) were collected at multiple
time points after injection of18F-FB-[Lys3]BBN (A) or18F-FB-
Aca-BBN(7–14) (B) (370 kBq/mouse). (C) Transaxial microPET
images of PC-3 tumor-bearing mice at 1 h after tail vein
injection of 3.7 MBq of18F-FB-[Lys3]BBN in absence (Control)
and presence (Block) of coinjected blocking dose of [Try4]BBN
(10 mg/kg mice body weight). Tumors are indicated by white
arrows in all cases.
microPET images of athymic nude mice bearing
18F-FB-Aca-BBN(7–14) (B) derived from 60-min dynamic micro-
PET scans. ROIs are shown for PC-3 tumor, liver, muscle, and
Time–activity curves of18F-FB-[Lys3]BBN (A) and
18F-BOMBESIN PET FOR GRPR • Zhang et al.497
with the tumor delineation. The presence of the well-
established tumor grown in the prostate gland was con-
firmed by microCTwithout a contrast agent (Fig. 7A). Good
visual agreement after registration was obtained in all
sagittal, coronal, and transaxial images (Fig. 7A). The reg-
istration is focused on the tumor region and did not use
markers that can be shifted or displaced. The whole reg-
istration procedure took ,15 min on a standard personal
computer, as the narrow-band approach used is a compact
representation of a structure where only pixels close to the
structure boundaries are considered (30). Both subcutane-
ous and orthotopic PC-3 tumor tissues were also resected
for histology to verify the characterization of tumors ex
vivo. The hematoxylin–eosin staining results (Fig. 7B) of
both PC-3 tumors showed similar morphology characteris-
tic of cancer cells.
There has been an exponential growth in the development
of radiolabeled peptides for diagnostic and therapeutic
applications in the last decade. Peptidic radiopharmaceuti-
cals have many favorable properties, including fast clear-
ance, rapid tissue penetration, and low antigenecity, and can
be produced easily and inexpensively (31). However, there
may be problems with the in vivo catabolism, unwanted
physiologic effects, and chelate attachment. Most studies
have been focused on radiometal labeling of peptides for
SPECT imaging of receptors that are overexpressed on the
diseased cells (32–34). More recently, peptides have been
conjugated to macrocyclic chelators for labeling of64Cu,
86Y, and68Ga for PET applications (17,20,35,36). Because
of the overexpression of GRPR in a variety of cancers,
bombesin analogs—derived either from the full tetradeca-
peptide sequence or from a truncated C-terminal portion of
the peptide—have been labeled with various radiometals
liver, kidney, and tumor homogenates collected at 1 h after
injection of18F-FB-[Lys3]BBN to a male athymic PC-3 tumor-
bearing nude mouse. As a comparison, the HPLC profile of
intact tracer (Standard) is also shown.
HPLC profiles of soluble fractions of blood, urine,
Extraction Efficiency and Elution Efficiency Data and HPLC Analysis of Soluble Fraction of Tissue Samples
at 60 Minutes After Injection
Extraction efficiency* (%)
FractionBlood UrineLiver KidneyPC-3
Elution efficiency (%)
HPLC analysis (%)
Intact tracer19.7 0.78.4 3.2 47.2
*Results in parentheses are from direct mixing of18F-FB-[Lys3]BBN with tissue homogenates.
yAmount of activity retained in pellets.
zAmount of activity that was extracted to PBS solution.
§Amount of activity that could not be trapped onto C18cartridge.
kAmount of activity that was eluted from C18cartridge using 2 mL of water.
¶Amount of activity that was eluted from C18cartridge using 2 mL of acetonitrile with 0.1% TFA.
ND 5 not determined.
498THE JOURNAL OF NUCLEAR MEDICINE • Vol. 47 • No. 3 • March 2006
for both PET (64Cu and68Ga) and SPECT (99mTc and111In)
imaging applications (14,15,17,18,20,37).18F is an ideal
short-lived PET isotope for labeling small molecular
recognition units, such as biologically active peptides,
and it is easily produced in the small biomedical cyclotrons.
Most peptides have the N-terminal primary amine group
and one or more lysine e-amino residues that can be labeled
group such as18F-SFB (22). Thus, we decided to label
both peptides ([Lys3]BBN and Aca-BBN(7–14)) with18F
for in vitro and in vivo characterizations.
Our cell-binding assay experiment demonstrated that the
truncated peptide sequence Aca-BBN(7–14) had significantly
lower receptor-binding affinity than that of [Lys3]BBN.
18F-labeled Aca-BBN(7–14) was also less potent than the
corresponding bombesin peptide analogs. Both tracers are
metabolically unstable after intravenous administration.
Multiple metabolites were found but not characterized here.
Identification of the composition of the degradation products
may be important to identify the cleavage sites to design and
characterize peptides of enhanced metabolic stability.
The internalization and efflux patterns of18F-FB-[Lys3]
18F-FB-Aca-BBN(7–14) are of note.
[Lys3]BBN with higher receptor affinity than18F-FB-Aca-
BBN(7–14) showed significantly higher cellular uptake.
Both tracers, however, had a rapid washout after reaching a
maximum at 15 min of incubation with PC-3 cells (Fig.
2A), which is similar to125I-[Tyr4]BBN but very different
from radiometal-labeled BBNs. The prolonged retention of
99mTc-,111In-, or64Cu-labeled BBNs is most likely due to
the lack of cell permeability of the hydrophilic macrocyclic
conjugate (14,15,17). In the case of18F-labeled bombesin
analogs, after GRPR-mediated internalization, both the intact
tracer and the metabolized peptide fractions that are radio-
active remain to be lipophilic and, thus, more amenable to
penetration in and out of the cells. It is, thus, not surprising
to observe rapid externalization of both18F-FB-[Lys3]BBN
BBN(7–14) effluxed even more rapidly than
[Lys3]BBN (Fig. 2B). Such in vitro characters of18F-labeled
bombesin analogs tallywiththe relativelyshorthalf-life of18F.
18F-FB-[Lys3]BBN with higher receptor affinity and pro-
longed cell retention than18F-FB-Aca-BBN(7–14) also exhib-
ited superior tumor-targeting efficacy and pharmacokinetics
in vivo. Although18F-FB-Aca-BBN(7–14) showed some
contrast at early time points, the activity accumulation in
the tumor was quickly washed out. Because of the lipophilic
18F-FB-Aca-BBN(7–14), it exhibited both
hepatobiliary and renal clearance routes as evidenced by
very strong signal in the liver, gallbladder, and lower abdo-
urinary bladder. A strong tumor-to-background contrast was
magnitude of tracer uptake is significantly lower than that
may have caused partial self-inhibition of receptor-specific
uptake in PC-3 tumor. We also noticed that nonradioactive
18F through an amine-reactive prosthetic labeling
topic PC-3 tumor. Representative transverse, coronal, and
sagittal images that contain the tumor at 17 min after injection
of 3.7 MBq of18F-FB-[Lys3]BBN are shown. The tumor grown in
mouse prostate gland is confirmed by microCT scan without
contrast agent. Coregistration of microPET (slice thickness, 1.2
mm) and microCT (slice thickness, 80 mm) is accomplished by
using a narrow-band approach without the need for fiducial
markers. (B) Hematoxylin–eosin staining (·400) of subcutane-
ous (left) and orthotopic (right) PC-3 tumor tissues.
(A) microPET and microCT visualization of ortho-
18F-BOMBESIN PET FOR GRPR • Zhang et al. 499
the tracer. The substantial blockade of tumor uptake by
unlabeled BBN suggests that some of the degraded radioac-
tive components accumulated in the tumor may also have
affinity for GRP receptor, which can be replaced by BBN.
[Lys3]BBN is a powerful tool for orthotopic prostate cancer
imaging. The high-resolution microCT scan provides good
contrast for PC-3 tumor without the need of contrast-
enhancing media, whereas microPET imaging with18F-
FB-[Lys3]BBN offers the GRPR expression level of the
tumor. In general, image registration can be formulated as
an optimization problem where the variables are a group of
transformation parameters that lead to the best match
between the input images. The match is quantified in
mathematic terms by the use of a metric, which ranks a
potential matching based on the image histograms, resolu-
tion, or pixel values of the involved organs. Usage of MI
has been widely adopted when dealing with multimodality
image registration (38). However, MI cannot be applied
directly to PET/CT registration for soft tissue because
the wide pixel intensities within an organ as imaged in
the PET images produce multiple correspondences with the
CT images that act as noise to the registration algorithm,
hindering its convergence (39). Therefore, only marker-
based techniques have been reported for PET/CT registra-
tion of mice studies (40). The narrow-band approach used
in this study was originally devised for magnetic resonance
spectroscopic imaging (MRSI)/CT registration, where a
similar noncorrelation of pixel intensities was observed
(30). Previous studies have suggested that this 2-step image
registration technique improves the convergence behavior
of the calculation and reduces the computational efforts
because sophisticated statistical considerations can be
replaced with simpler pixel-based metrics computed only
in the regions of clinical interest.
This study demonstrated the successful coupling of
[Lys3]BBN and Aca-BBN(7–14) with positron-emitting
18F through the prosthetic labeling group
18F-SFB. The bombesin analog with the full tetradecapep-
tide sequence (18F-FB-[Lys3]BBN) is superior to that with a
truncated C-terminal sequence (18F-FB-Aca-BBN(7–14))
in terms of GRPR avidity, receptor-mediated internalization
rate, intracellular retention, tumor-targeting efficacy, and
in vivo pharmacokinetics. Although18F-FB-[Lys3]BBN is
relatively metabolically unstable, dynamic PET scans dem-
onstrated the ability of this tracer to visualize both subcutane-
ous and orthotopic PC-3 tumor in murine xenograft models.
Furthermore,18F-FB-[Lys3]BBN may also be used for local-
ization of other tumors that are GRPR positive.
This work was supported, in part, by Department of
Defense (DOD) Prostate Cancer Research Program (PCRP)
New Investigator Award (NIA) DAMD1717-03-1-0143,
National Cancer Institute (NCI) grant R21 CA102123,
National Institute of Biomedical Imaging and Bioengineer-
ing (NIBIB) grant R21 EB001785, DOD Breast Cancer
Research Program (BCRP) Concept Award DAMD17-03-
1-0752, DOD BCRP IDEA Award W81XWH-04-1-0697,
American Lung Association California, Society of Nuclear
Medicine Education and Research Foundation, NCI Small
Animal Imaging Resource Program (SAIRP) grant R24
CA93862, and NCI In Vivo Cellular Molecular Imaging
Center (ICMIC) grant P50 CA114747. Dr. Zhengming
Xiong is acknowledged for cell culture and the authors
thank Pauline Chu for histology.
1. di Sant’Agnese PA. Neuroendocrine cells of the prostate and neuroendocrine
differentiation in prostatic carcinoma: a review of morphologic aspects. Urology.
2. Vashchenko N, Abrahamsson PA. Neuroendocrine differentiation in prostate
cancer: implications for new treatment modalities. Eur Urol. 2005;47:147–
3. Chung DH, Evers BM, Beauchamp RD, et al. Bombesin stimulates growth of
human gastrinoma. Surgery. 1992;112:1059–1065.
4. Glover SC, Tretiakova MS, Carroll RE, Benya RV. Increased frequency of
gastrin-releasing peptide receptor gene mutations during colon-adenocarcinoma
progression. Mol Carcinog. 2003;37:5–15.
5. Battey J, Wada E, Corjay M, et al. Molecular genetic analysis of two distinct
receptors for mammalian bombesin-like peptides. J Natl Cancer Inst Monogr.
6. Scheffel U, Pomper MG. PET imaging of GRP receptor expression in prostate
cancer. J Nucl Med. 2004;45:1277–1278.
7. Matusiak D, Glover S, Nathaniel R, Matkowskyj K, Yang J, Benya RV.
Neuromedin B and its receptor are mitogens in both normal and malignant
epithelial cells lining the colon. Am J Physiol Gastrointest Liver Physiol. 2005;
8. Fathi Z, Corjay MH, Shapira H, et al. BRS-3: a novel bombesin receptor subtype
selectively expressed in testis and lung carcinoma cells. J Biol Chem. 1993;
9. Pinski J, Reile H, Halmos G, Groot K, Schally AV. Inhibitory effects of
somatostatin analogue RC-160 and bombesin/gastrin-releasing peptide antago-
nist RC-3095 on the growth of the androgen-independent Dunning R-3327-AT-
1 rat prostate cancer. Cancer Res. 1994;54:169–174.
10. Reubi JC, Macke HR, Krenning EP. Candidates for peptide receptor radiotherapy
today and in the future. J Nucl Med. 2005;46(suppl 1):67S–75S.
11. Maina T, Nock BA, Zhang H, et al. Species differences of bombesin analog
interactions with GRP-R define the choice of animal models in the development
of GRP-R-targeting drugs. J Nucl Med. 2005;46:823–830.
12. Varvarigou A, Bouziotis P, Zikos C, Scopinaro F, De Vincentis G. Gastrin-
releasing peptide (GRP) analogues for cancer imaging. Cancer Biother
13. Van de Wiele C, Dumont F, Dierckx RA, et al. Biodistribution and dosimetry of
99mTc-RP527, a gastrin-releasing peptide (GRP) agonist for the visualization of
GRP receptor-expressing malignancies. J Nucl Med. 2001;42:1722–1727.
14. Scopinaro F, De Vincentis G, Corazziari E, et al. Detection of colon cancer with
99mTc-labeled bombesin derivative (99mTc-Leu13-BN1). Cancer Biother Radio-
15. Zhang H, Chen J, Waldherr C, et al. Synthesis and evaluation of bombesin
derivatives on the basis of pan-bombesin peptides labeled with indium-111,
lutetium-177, and yttrium-90 for targeting bombesin receptor-expressing tumors.
Cancer Res. 2004;64:6707–6715.
16. Smith CJ, Sieckman GL, Owen NK, et al. Radiochemical investigations of
[188Re(H2O)(CO)3-diaminopropionic acid-SSS-bombesin(7-14)NH2]: syntheses,
radiolabeling and in vitro/in vivo GRP receptor targeting studies. Anticancer Res.
17. Rogers BE, Bigott HM, McCarthy DW, et al. MicroPET imaging of a gastrin-
releasing peptide receptor-positive tumor in a mouse model of human prostate
cancer using a64Cu-labeled bombesin analogue. Bioconjug Chem. 2003;14:
500THE JOURNAL OF NUCLEAR MEDICINE • Vol. 47 • No. 3 • March 2006
18. Rogers BE, Manna DD, Safavy A. In vitro and in vivo evaluation of a64Cu- Download full-text
labeled polyethylene glycol-bombesin conjugate. Cancer Biother Radiopharm.
19. Chen X, Park R, Hou Y, et al. MicroPET and autoradiographic imaging of GRP
receptor expression with
64Cu-DOTA-[Lys3]bombesin in human prostate
adenocarcinoma xenografts. J Nucl Med. 2004;45:1390–1397.
20. Schuhmacher J, Zhang H, Doll J, et al. GRP receptor-targeted PET of a rat
pancreas carcinoma xenograft in nude mice with a68Ga-labeled bombesin(6-14)
analog. J Nucl Med. 2005;46:691–699.
21. Okarvi SM. Recent progress in fluorine-18 labelled peptide radiopharmaceuti-
cals. Eur J Nucl Med. 2001;28:929–938.
22. Vaidyanathan G, Zalutsky MR. Improved synthesis of N-succinimidyl
4-[18F]fluorobenzoate and its application to the labeling of a monoclonal
antibody fragment. Bioconjug Chem. 1994;5:352–356.
23. Chen X, Park R, Shahinian AH, et al.18F-Labeled RGD peptide: initial evaluation
for imaging brain tumor angiogenesis. Nucl Med Biol. 2004;31:179–189.
24. Wu Y, Zhang X, Xiong Z, et al. MicroPET imaging of glioma av-integrin expres-
sion using64Cu-labeled tetrameric RGD peptide. J Nucl Med. 2005;46:1707–1718.
25. Chen X, Park R, Hou Y, et al. MicroPET imaging of brain tumor angiogenesis
with18F-labeled PEGylated RGD peptide. Eur J Nucl Med Mol Imaging. 2004;
26. Chen X, Tohme M, Park R, Hou Y, Bading JR, Conti PS. Micro-PET imaging of
avb3-integrin expression with18F-labeled dimeric RGD peptide. Mol Imaging.
27. Chen X, Park R, Tohme M, Shahinian AH, Bading JR, Conti PS. MicroPET and
autoradiographic imaging of breast cancer av-integrin expression using18F- and
64Cu-labeled RGD peptide. Bioconjug Chem. 2004;15:41–49.
28. Hillman GG, Wang Y, Kucuk O, et al. Genistein potentiates inhibition of tumor
growth by radiation in a prostate cancer orthotopic model. Mol Cancer Ther.
29. Boellaard R, van Lingen A, Lammertsma AA. Experimental and clinical
evaluation of iterative reconstruction (OSEM) in dynamic PET: quantitative
characteristics and effects on kinetic modeling. J Nucl Med. 2001;42:808–817.
30. Schreibmann E, Xing L. Narrow band deformable registration of prostate mag-
netic resonance imaging, magnetic resonance spectroscopic imaging, and com-
puted tomography studies. Int J Radiat Oncol Biol Phys. 2005;62:595–605.
31. Benedetti E, Morelli G, Accardo A, Mansi R, Tesauro D, Aloj L. Criteria for the
design and biological characterization of radiolabeled peptide-based pharma-
ceuticals. BioDrugs. 2004;18:279–295.
32. Nock BA, Nikolopoulou A, Galanis A, et al. Potent bombesin-like peptides for
GRP-receptor targeting of tumors with99mTc: a preclinical study. J Med Chem.
33. Janssen ML, Oyen WJ, Dijkgraaf I, et al. Tumor targeting with radiolabeled avb3
integrin binding peptides in a nude mouse model. Cancer Res. 2002;62:6146–
34. De Jong M, Valkema R, Van Gameren A, et al. Inhomogeneous localization of
radioactivity in the human kidney after injection of [111In-DTPA]octreotide.
J Nucl Med. 2004;45:1168–1171.
35. Chen X, Liu S, Hou Y, et al. MicroPET imaging of breast cancer av-integrin
expression with64Cu-labeled dimeric RGD peptides. Mol Imaging Biol. 2004;
36. Chen X, Sievers E, Hou Y, et al. Integrin avb3-targeted imaging of lung cancer.
37. Hoffman TJ, Gali H, Smith CJ, et al. Novel series of111In-labeled bombesin
analogs as potential radiopharmaceuticals for specific targeting of gastrin-
releasing peptide receptors expressed on human prostate cancer cells. J Nucl
38. Maes F, Collignon A, Vandermeulen D, Marchal G, Suetens P. Multimodality
image registration by maximization of mutual information. IEEE Trans Med
39. du Bois d’Aische A, Craene MD, Geets X, Gregoire V, Macq B, Warfield SK.
Efficient multi-modal dense field non-rigid registration: alignment of histological
and section images. Med Image Anal. 2005;9:538–546.
40. Thomas CT, Meyer CR, Koeppe RA, et al. A positron-emitting internal marker
for identification of normal tissue by positron emission tomography: phantom
studies and validation in patients. Mol Imaging Biol. 2003;5:79–85.
18F-BOMBESIN PET FOR GRPR • Zhang et al. 501