Abstract— This study aims to evaluate the in vivo
distribution of Gold Nanoparticles (GNPs) at different time
points after intratumoral (IT) injection, exploiting their
properties as contrast agents for Computed Tomography
(CT). GNPs approximately 40 nm in diameter were
synthesized with a surface plasmon peak at ~530 nm, capped
with Bovine Serum Albumin (BSA) to improve colloidal
stability, and characterized with standard methods. CT
phantom imaging was performed to quantify X-ray
attenuation as a function of GNP concentration and surface
functionalization and to determine the appropriate particle
dose for in vivo studies. Concentrated GNPs were
intratumorally (IT) injected into Lewis Lung Carcinoma
(LLC) solid tumors grown on the right flank of 6-week old
female C57BL/6 mice. Ten days post-injection, follow up CT
imaging was performed to assess the distribution and
retention of the particles in the tumor. Using the CT
attenuation quantification, images for each timepoint were
segmented, and 3D volumes rendered, to conduct
biodistribution analyses. The successful retention and
permanence of the GNPs into the solid tumor after ten days
suggests the significance of GNPs as a potential theranostic
Non-small cell lung cancer (NSCLC) ranks among the
most common type of lung cancer, accounting for ∼85% of
all lung cancer diagnoses, with 5-year cause-specific
survival rates ranging from 13% to 32% and local failure
rates of 42% to 49% when treated with conventional
radiotherapy alone . The engagement to reduce the
radiation dose and damage to healthy tissues without losing
efficacy in cancer therapy have focused the research on
exploiting gold nanoparticle (GNP) properties in enhancing
radiation effects via physical, chemical, and biological
interactions with ionizing radiation . However,
challenges concerning in vivo efficacy still limit the clinical
translation of GNP radiosensitizers mainly due to the lack of
colloidal stability, clearance, and possible long-term toxicity
. Therefore, the ability to predict and determine GNP in
vivo biodistribution will provide further information about
the preferred uptake pathways for achieving precise dose
deposition as well as a better understanding of mechanisms
behind radiosensitization . To quantify the biodistribution
*Research supported by the Simmons Foundation, Houston Methodist
Research Institute (CF), and Golfers Against Cancer.
R.T. is with the Department of Nanomedicine, Houston Methodist
Research Institute, TX, & Dept. of Electronics, Politecnico di Torino, Italy
E.B.B. is with the Department of Radiation Oncology, Houston
Methodist Research Institute, TX. D.D. is with the Dept. of Electronics,
Politecnico di Torino, Italy.
of GNPs within cells and tissue, several imaging techniques
can be employed due to the physical properties of GNPs that
allow them to act as imaging agents , . This study
focuses on Computed Tomography (CT) as an imaging
modality to provide insight into the GNP local
biodistribution in Lewis Lung Carcinoma (LLC) tumor-
bearing mice based on CT attenuation levels. CT is an
inexpensive diagnostic imaging system routinely used in
practice with a high capability of deeply tissue permeation
and density resolution which allows for 3-dimensional (3D)
reconstructions of X-ray images. CT contrast agents are
commonly administered to improve the contrast among
tissues with similar or lower attenuation values by
increasing the signal-to-noise ratio without additional
radiation doses to the patient. Nevertheless, different
drawbacks in the use of iodine-based contrast agents such as
renal toxicity, deficiency in the targeting process, and
insufficient circulation time as well as amplification in DNA
damage during CT scans, restrict their application .
Several studies confirm GNPs as excellent contrast
agents for CT imaging and multimodalities –. So far,
CT image-guided cancer treatments using radio-enhancing
GNPs has not been thoroughly investigated in preclinical
NSCLC models. The main contribution of this study is to
give a better understanding of the in vivo IT distribution and
retention of citrate-capped and BSA-capped GNPs using CT
A. GNPs Synthesis and characterization
Multi-faceted GNPs (Figure 1A) are synthesized using
an adapted protocol by Turkevich et al.  to tune the
particle size to a value of approximately 40 nm. Briefly, 1
g of gold chloride (AuCl3, purity > 99.99%) is added to 100
mL of Milli-Q water and filtered through a 0.22 μm nylon
filter, and 1 g of citric acid (C6H8O7) was dissolved in 100
mL of Milli-Q under stirring.
Respectively, 4.8 mL of the 1% (w/v) citric acid solution
and 7 mL of 1% (w/v) gold chloride premade reagents are
added to 600 mL of boiling Milli-Q water. The synthesis is
M.S., D.W., S.R., M.H., N.F. are with the Department of
Nanomedicine, Houston Methodist Research Institute, TX.
A.G. is with the Department of Nanomedicine, Department of
Radiation Oncology and Department of Surgery Houston Methodist
Research Institute, TX.
C.S.F. is with the Department of Nanomedicine and Department of
Cardiovascular Surgery, Houston Methodist Research Institute, TX,
(phone: 713-441-1996; email: firstname.lastname@example.org)
Intratumoral Gold Nanoparticle-Enhanced CT Imaging: An in Vivo
Investigation of Biodistribution and Retention*
Rossana Terracciano, Marc L. Sprouse, Dennis Wang, Sara Ricchetti, Matteo Hirsch, Nicola Ferrante,
E. Brian Butler, Danilo Demarchi, Alessandro Grattoni and Carly S. Filgueira
complete when the color has changed from black to dark
The GNPs are characterized with UV-Vis Spectroscopy
and electron microscopy (SEM, TEM). Dynamic Light
Scattering (DLS) was used to rapidly and qualitatively size
the particles and obtain a polydispersity index (PDI). Zeta
Potential was also measured (Malvern).
The resulting particles yielded a surface plasmon peak at
~530 nm with a hydrodynamic diameter of 43.89 ± 15.45
nm (Figure 1B). The particle diameter extracted from SEM
and TEM images is 39.1 ± 15.0 nm, which is in accordance
with the value obtained with DLS (< 11% error). The
particle solutions yield Z-potentials of -40.0 ± 6.0 mV.
B. BSA capping and characterization
The GNP surface is then modified by passive absorption
of Bovine Serum Albumin (BSA) to increase colloidal
stability and biocompatibility. To perform BSA-capping on
the GNP surface, lyophilized BSA in a concentration of 2%
(w/v) is added to the solution of colloidal gold already
prepared and characterized. The mixture is stirred
vigorously until complete dissolution of BSA. The
absorption of BSA on the particle surface is evaluated by
analyzing the UV-Vis spectrum of the solution. A positive
passivation of the particles is confirmed by a red-shift in the
gold SPR peak at 535 nm (Figure 1b) and the presence of a
BSA absorption peak at 280 nm.
Figure 1. Size distribution (n>100) and TEM image of citrate-GNPs (A).
Average particle size: 39.1 ± 15.0 nm. Absorbance spectra of citrate-
capped GNPs and BSA-GNPs with absorbance maxima occurring at
530nm and 535nm, respectively (B).
C. CT phantom imaging
In order to prepare the samples for clinical CT phantom
imaging, GNP solutions in concentrations ranging from 0
to 10mg/mL are aliquoted in 100 µL tubes and scanned with
a Siemens Inveon High-Resolution CT to assess the CT
contrast properties (Figure 2). The phantom experiment was
carried out using CT parameters with slice thickness of 105
µm, in plane resolution of 105 µm, tube voltage at 80 kV,
tube current at 500 µA, exposure time of 240 ms, and by
placing the samples directly on the animal bed.
X-ray attenuation intensity was determined in
Hounsfield unit (HU) by processing the digital CT images
(DICOM files) using 3DSlicer and selecting a 3D
reconstructed region of interest (ROI) for each sample and
then recording the mean attenuation value and plotting as a
function of gold and/or iodine concentration in mg/mL.
As expected, an increase of CT attenuation occurred
when the mass concentration of the GNPs increased. Both
citrate-GNPs and BSA-GNPs absorb more X-rays than
Omnipaque350 (a standard iodine-based CT contrast agent)
as shown in Figure 3. The range of CT values for LLC-LUC
solid tumor 8 days after cell inoculation is included in the
graph for comparison.
D. In vivo CT imaging
High-resolution CT imaging is then exploited to compare
the biodistribution of citrate and BSA-capped GNPs in a
LLC murine model of NSCLC. In vivo imaging
experiments were performed using six-week-old female
C57/BL6 mice, purchased from Taconic Bioscence
(Rensselaer, NY, USA).
The mice received an injection of 2x106 LLC cells in the
right flank subcutaneously once their weight reached an
average of 18.4 g. After approximately 10 days post-
injection, the volumes of tumor nodules appeared spherical
and 100 mm3 in volume. The mice were anesthetized using
isoflurane, and 100 μL of citrate or BSA-capped GNPs
were injected IT (3.5 mg/mL of gold). CT images pre-
injection were recorded as a baseline for the biodistribution
analysis. All the CT imaging was performed using a
Siemens Inveon Multi-Modality (MM) System controlled
with the Inveon Acquisition Workplace (IAW). In vivo
experiment was carried out using CT parameters with slice
thickness of 103.25 µm, in plane resolution of 103.25 µm,
tube voltage at 80 kV, tube current at 500 µA, and exposure
time of 240 ms.
The mice were imaged at different time intervals (day 0
pre-injection, day 0 post-injection, day 3 post-injection, day
6 post-injection, day 9 post-injection, and day 10 post-
injection). The CT images were acquired and reconstructed
with 3DSlicer software.
III. DATA ANALYSIS
Figure 4A shows the 3D renderings of three significant
timepoints during the ten day post-injection time period.
Although both citrate and BSA-capped GNPs exhibited
excellent biocompatibility and sustained retention post IT
injection, the citrate-GNPs were observed to stably cluster
in situ over the ten days, while the BSA-GNPs were less
likely to locally agglomerate within the tumor (Figure 4B,
Figure 4C). Therefore, the BSA helps to facilitate the
biodistribution of the GNPs in the tumor area, preventing
the formation of clusters.
This evidence is also supported by Figure 5, which
shows the differences in particle distribution volume within
the tumor tissue. On day 0, the 3D reconstructed volume of
the citrate-GNPs from the attenuation values extracted from
the CT images is four times smaller than those created by
the BSA-GNPs injection. 72h after particle injection, both
the citrate and BSA-capped GNPs follow the same constant
trend in volume until the tenth day.
Figure 2. CT Contrast Proprieties of ~40 nm citrate and BSA-capped GNPs
Compared with those of Omnipaque350. Photos of the phantom dilution
series from 0 (water) to 10 mg/mL concentrations for Omnipaque350
(blue), citrate-GNPs (green), and 2% (w/v) BSA-capped GNPs (red) (A).
Representative 3D volume rendered CT images and sample phantom
images from Siemens Inveon High Resolution CT scanner (B). For a better
understanding and comparison is reported a ROI of the LLC tumor 8 days
after cell inoculation and immediately before GNP injection (C).
In order to better understand the IT biodistribution pattern
over time and eventually the clearance of the particles, we
quantified the contrast in several organs at different time
points. Figure 6 represents the CT values of attenuation
(HU) of the heart, brain, kidneys, liver, intestine, tumor and
bladder, extracted from circular ROIs selected on the CT
images at the different time points after administration of
citrate and BSA-capped GNPs. Due to the presence of
clusters in the case of injections of citrate-GNPs, the
analysis is divided into sub-measures for the tumor:
1. GNPs Intra-cluster (IC) to indicate the ROI inside
the tumor and inside the cluster of GNPs
2. GNPs Extra-cluster (EC) to indicate the ROI
inside the tumor and outside the cluster of GNPs.
0 2 4 6 8 10
2% (w/v) BSA-GNPs
Y = 33.17x + 1.790
R2 = 0.99
Y = 16.53x + 10.67
R2 = 0.98
Y = 31.04x + 1.416
R2 = 0.98
Figure 3. X-ray attenuation changes in Hounsfield Units (HU) versus
concentration for citrate and BSA-capped GNPs compared with those of
Omnipaque350. Data are reported in terms of mean value of the 3D
reconstructed voxel attenuations and standard deviations. To trace the
thresholds for choosing the optimal concentration to inject, the LLC tumor
tissue density range 8 days after cell inoculation and immediately before
GNPs injection is reported (yellow range).
As seen in Figure 6A, citrate-GNPs clusters are well
enclosed in the tumor and remain there even after 10 days.
The extra-cluster tumor area doesn’t show the presence of
particles over time. The absence of citrate-GNPs over time
in this area indicates a biodistribution capability of the
particles under the limits of detection. The CT attenuation
results here suggest that the citrate-GNPs are not excreted
in 10 days. This conclusion is further deduced because
when compared to the baseline, no visible attenuation
increments are distinguishable in other organs over time.
Conversely, the CT results from the images of BSA-
GNPs-injected mice confirm contrast enhancement in the
tumor ROI at the first time point (immediately after
injection of BSA-GNP) with attenuation values of 354 HU,
~8% higher than the baseline (325 HU). Furthermore, the
contrast remains stable over the subsequent 3 days, while at
day 10, comes back to baseline levels. Despite the high
accumulation in the intestine and partial changes in the
other organs (heart, brain, kidneys, liver and bladder), the
results suggest that the particles could be excreted after 10
days. Heart, brain, kidneys and liver all returned to baseline
levels after 10 days, while the CT signals of the intestine
increased gradually, indicating bile excretion of the
particles. Accumulation in the bladder is still present after
10 days. This could indicate a nearly complete excretion of
the particles from the body after 10 days.
Despite their good biodistribution capabilities, a point of
discussion is the lack of in vivo contrast enhancement of the
BSA-GNPs when compared to the citrate-GNPs (~50% and
~8%, respectively). It becomes clear that a better and more
immediate distribution in the tumor environment
compromises CT contrast properties. Hence, the
importance of calibrating the CT attenuation of the GNPs
before in vivo injection for tracing a safe threshold and
leveraging the CT-contrast proprieties of the particles.
Figure 4. Representative 3D volumes rendered CT images at day 0
immediately after IT injections of 100µL BSA-capped GNPs and citrate-
GNPs solutions, day 6 and day 10 after IT injection. All the images are
displayed at window width of 1500 HU and window level 700 to cover
bone, metal and tumor tissue attenuations. The voxel size of the images is
105µm. The grayscale look up table is then shown in a more natural-like
color and GNPs are displayed in yellow (A). CT slices of mice bearing
tumors, cropped specifically in the tumor region immediately after BSA-
GNPs injection (B) and citrate GNPs (C), where the clustering process is
Figure 5. 3D reconstructed volumes (mm3) of the GNPs from the
attenuation values extracted from the CT images over the imaging
Figure 6. The CT attenuation values (HU) of various organs at different
time points after administration of (A) citrate-GNPs and (B) BSA-GNPs.
The analysis for citrate-GNPs divides the tumor ROIs in GNPs Intracluster
(IC) and GNPs Extracluster (EC).
In conclusion, the permanence of the GNPs into the
tumor after ten days suggests the significance of GNPs as a
potential theranostic agent. Further, the more homogeneous
and uniform IT distribution of the BSA-GNPs may offer
further advantages for surface passivation.
All experiments conducted on mice were approved by the
Institutional Animal Care and Use Committee (IACUC) at
Houston Methodist Research Institute and were performed
according to the principles of the NIH Guide for the Care
and Use of Laboratory Animals, the provisions of the
Animal Welfare Act, PHS Animal Welfare Policy, and the
policies of the Houston Methodist Research Institute.
Housing and care were provided in accordance with the
regulations of the Animal Welfare Act and
recommendations of the Guide for the Care and Use of
0 2 4 6 8 10 12
Days Post GNPs Injection
CT Extracted GNPs Volume
Tumor (GNPs IC)
Tumor (GNPs EC)
Day 0 citrate-GNPs
Day 0 citrate-GNPs
Day 6 citrate-GNPs
Day 10 citrate-GNPs
Day 0 BSA-GNPs Pre-Injection
Day 0 BSA-GNPs Post-Injection
Day 3 BSA-GNPs
Day 6 BSA-GNPs
Day 9 BSA-GNPs
Day 10 BSA-GNPs
Funding support was received from the Simmons
Foundation, Houston Methodist Research Institute (CF),
and Golfers Against Cancer agencies. We are grateful to Dr.
Xukui Wang, the Houston Methodist Research Institute
Translational Imaging - PreClinical Imaging (Small
Animal) Core, and the Houston Methodist Research
Institute Microscopy Core.
 C. J. Miller et al., “Predictors of Distant Failure After Stereotactic
Body Radiation Therapy for Stages I to IIA Non–Small-Cell Lung
Cancer,” Clin. Lung Cancer, vol. 20, no. 1, pp. 37–42, Jan. 2019,
 S. Her, D. A. Jaffray, and C. Allen, “Gold nanoparticles for
applications in cancer radiotherapy: Mechanisms and recent
advancements,” Adv. Drug Deliv. Rev., vol. 109, pp. 84–101, Jan.
2017, doi: 10.1016/j.addr.2015.12.012.
 Y. Chen et al., “Gold nanoparticles coated with polysarcosine
brushes to enhance their colloidal stability and circulation time in
vivo,” J. Colloid Interface Sci., vol. 483, pp. 201–210, Dec. 2016,
 S. Rosa, C. Connolly, G. Schettino, K. T. Butterworth, and K. M.
Prise, “Biological mechanisms of gold nanoparticle
radiosensitization,” Cancer Nanotechnol., vol. 8, no. 1, p. 2, Dec.
2017, doi: 10.1186/s12645-017-0026-0.
 D. Kim et al., “Label-free high-resolution 3-D imaging of gold
nanoparticles inside live cells using optical diffraction
tomography,” Methods, vol. 136, pp. 160–167, Mar. 2018, doi:
 R. Meir et al., “Nanomedicine for Cancer Immunotherapy:
Tracking Cancer-Specific T-Cells in Vivo with Gold Nanoparticles
and CT Imaging,” ACS Nano, vol. 9, no. 6, pp. 6363–6372, Jun.
2015, doi: 10.1021/acsnano.5b01939.
 M. M. Mahan and A. L. Doiron, “Gold Nanoparticles as X-Ray,
CT, and Multimodal Imaging Contrast Agents: Formulation,
Targeting, and Methodology,” Journal of Nanomaterials, 2018.
Mar. 23, 2020).
 R. Cheheltani et al., “Tunable, biodegradable gold nanoparticles as
contrast agents for computed tomography and photoacoustic
imaging,” Biomaterials, vol. 102, pp. 87–97, Sep. 2016, doi:
 M. F. Ghaziyani et al., “Anti-CD24 bio Modified PEGylated Gold
Nanoparticles as Targeted Computed Tomography Contrast
Agent,” Adv. Pharm. Bull., vol. 8, no. 4, pp. 599–607, Nov. 2018,
 Y. C. Dong et al., “Effect of Gold Nanoparticle Size on Their
Properties as Contrast Agents for Computed Tomography,” Sci.
Rep., vol. 9, no. 1, pp. 1–13, Oct. 2019, doi: 10.1038/s41598-019-
 J. Turkevich, P. C. Stevenson, and J. Hillier, “The Formation of
Colloidal Gold,” J. Phys. Chem., vol. 57, no. 7, pp. 670–673, Jul.
1953, doi: 10.1021/j150508a015.