Journal of Controlled Release 374 (2024) 242–253

Contents lists available at ScienceDirect

Journal of Controlled Release

journal homepage: www.elsevier.com/locate/jconrel

Lipiodol emulsion as a dual chemoradiation-sensitizer for pancreatic

cancer treatment
Shuang Zhu a, c, 1 , Chenglu Gu a, b, 1 , Long Gao d, e, 1 , Shuanglong Du a, b , Duiping Feng d, e ,
Zhanjun Gu a, b, *
a
CAS Key Laboratory for Biomedical Effects of Nanomaterials and Nanosafety, Institute of High Energy Physics, Chinese Academy of Sciences, Beijing 100049, China
b
Center of Materials Science and Optoelectronics Engineering, College of Materials Science and Optoelectronic Technology, University of Chinese Academy of Sciences,
Beijing 100049, China
c
Spallation Neutron Source Science Center, Institute of High Energy Physics, Dongguan 523803, China
d
Shanxi Provincial Clinical Research Center for Interventional Medicine, First Hospital of Shanxi Medical University, Taiyuan 030001, China
e
Department of Oncological and Vascular Intervention, First Hospital of Shanxi Medical University, Taiyuan 030001, China

A R T I C L E I N F O A B S T R A C T

Keywords: Pancreatic ductal adenocarcinoma (PDAC) has a low survival rate and limited treatment options. Concurrent
Lipiodol emulsion chemoradiotherapy is considered beneficial to improve tumor control, but the low drug bioavailability at tumor
Pancreatic cancer site and the low radiation tolerance of surrounding healthy organs greatly limits its effectiveness. Lipiodol, a
Concurrent chemoradiotherapy
natural drug carrier used in clinical transarterial chemoembolization, has shown potential as a radiosensitizer
Radiosensitizer
Drug carrier
due to its high Z element iodine composition. Thus, this study aims to repurpose lipiodol as a sensitizer to
simultaneously enhance chemo- and radiotherapy for PDAC. To this end, a stable lipiodol emulsion (IOE) loaded
with gemcitabine is designed using clinically approved surfactants. At in vivo level, IOE demonstrates better
radiotherapeutic effect than existing nanoradiosensitizers and enhanced drug bioavailability over free drug,
leading to significant tumor inhibition and improved survival rates under concurrent chemo-radiotherapy. This
may due to the sustained drug release, homogenous spatial distribution, and long-term retention ability of IOE in
solid PDAC tumor. Furthermore, to better understand the functioning mechanism of drug-loaded IOE, in vitro
study is conducted to reveal the ROS- and DNA damage-related therapeutic pathways. Lastly, a comprehensive
toxicity assessment also proves the good biocompatibility and safety of as-prepared IOE. This study offers a
clinically feasible sensitizer for simultaneous chemoradiotherapy and holds potential for other types of cancer
treatment in clinics.

1. Introduction response to chemoradiotherapy. [3] This is because pancreatic cancers

are inherently resistant to these treatments. For chemotherapy, the
Pancreatic cancer, comprising mostly pancreatic ductal adenocarci- PDAC is characterized by an extremely fibrotic and abnormal vascula-
noma (PDAC), is the third leading cause of cancer death with the lowest ture. This could induce elevated interstitial fluid pressure and impaired
survival rate of 12% among all cancer types. [1,2] Surgery remains the intratumoral blood perfusion, and thus greatly hinder drug delivery and
top treatment option for early-stage disease, but over 80% of patients are infiltration upon systematic administration. [4] To tackle this issue,
usually diagnosed late in cancer progression. For patients with locally using intratumoral injection and a suitable drug carrier to increase the
advanced, unresectable tumor, concurrent chemotherapy and radio- selectivity and bioavailability of anticancer drug at tumor site is desir-
therapy (chemoradiotherapy) is considered beneficial to improve local able. For radiotherapy, due to the anatomical location, the key limita-
disease control. However, treatment resistance is still a major clinical tion for radiotherapy of PDAC is the low radiation tolerance of
challenge with only 4% of patients achieving pathological complete surrounding healthy organs. Therapeutic radiation doses usually cause

* Corresponding author at: CAS Key Laboratory for Biomedical Effects of Nanomaterials and Nanosafety, Institute of High Energy Physics, Chinese Academy of
Sciences, Beijing 100049, China.
E-mail address: zjgu@ihep.ac.cn (Z. Gu).
1
These authors contributed equally to this work.

https://doi.org/10.1016/j.jconrel.2024.08.020
Received 8 May 2024; Received in revised form 17 July 2024; Accepted 13 August 2024
Available online 20 August 2024
0168-3659/© 2024 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
S. Zhu et al. Journal of Controlled Release 374 (2024) 242–253

severe damage to surrounding radiosensitive organs like the duodenum the versatile drug delivery capacity, lipiodol emulsion holds a great
or gastrointestinal tract. [5] Although several molecular-targeted radi- potential to be a universal sensitizing system for multi-drug-based che-
osensitizers have been developed to increase the therapeutic effect of moradiotherapy in clinics.
radiotherapy [6], the efficacy of single pathway-based biological radi-
osensitizers is limited. This is due to the heterogeneity of PDAC, with its 2. Experiment Section
multiple subpopulations of tumor cells exhibiting diverse biological
traits and metabolic functions, further complicates treatment. [7] 2.1. Preparation of lipiodol emulsion
Recently, high Z element-based materials as a broad-spectrum radio-
sensitizer have been widely studied and even approved in clinics (such Lipiodol emulsion (IOE) is prepared by mixing oil phase, aqueous
as a HfO2-based radiosensitizer NBTXR3 (Nanobiotix SA, France)) for phase and emulsifier in a mechanical method. The surfactant concen-
enhancing radiotherapeutic effect. [8–10] In theory, high Z element tration is determined according to previous reports where Tween 80 is
materials can release electrons (such as photoelectric and Compton/ used between 1 and 15% and HCO 40 < 5% (V/V) [18]. The preparation
Auger electrons) upon interaction with incident photons, causing direct of emulsion follows the safety standard. In brief, 5.4 mL of lipiodol
cellular DNA damage or indirect reactive oxygen species (ROS) pro- (Guerbet), 3.6 mL of physiological saline (Kelun Pharmaceutical), 108
duction to promote cell death. [11] Once located in tumor cells, they can μL Tween 80 (Aladdin) and 432 of μL HCO 40 (Boer Chemical Reagent)
enhance the local deposition of radiation energy, thus boosting radiation are mixed. Then the mixture is sonicated for 5 min using an ultrasonic
therapy efficacy in the tumors while leaving adjacent healthy tissues cell Crusher (600 W, 3 s working/2 s interval, Ningbo Scientz Biotech-
unharmed. [12] Based on the physical sensitizing mechanism, high Z- nology Co.), and homogenized for 5 min (12,000 rpm) using a tissue
dependent radiosensitizer is considered to have a universal radio- homogenizer (Jingxin Industrial Development Co.) to form an oil-in-
sensitizing ability for any type of tumors, regardless of genetic makeup water emulsion (Video S1). For the preparation of IOD, gemcitabine is
or histology. From this point of view, developing a high Z-based drug first dissolved in physiological saline solution, and the final concentra-
carrier may be a feasible strategy to simultaneously sensitize chemo- tion of drug in emulsion is 35 mg/mL.
radiotherapy of PDAC for enhanced therapeutic outcome.
One of the most effective strategies in drug development is drug
repurposing as it can alleviate the required cost and timeline. [13] We 2.2. Particle size analysis
thus focus on looking for the chemoradiosensitizers from existing
medicines. During the transarterial chemoembolization (TACE) for he- 5 μL IOE is dissolved in 5 mL pure water, and then the solution is
patocellular carcinoma (HCC), lipiodol is widely used as a drug carrier diluted 100 times again, and placed 1 mL in the particle size analyzer
based on its tumor-seeking and -retention property. Upon suspended (DLS, Brookhaven) to detect the particle size and Zeta potential of the
into lipiodol, cytotoxic drug could be selectively injected and evenly dispersed emulsion. For particle size detection through syringe needles,
retained within tumor tissue to allow sustained impregnation via 5 μL IOE was injected into pure water via a 0.3 mL syringe (31G * 8 mm).
interventional method. [14] This could greatly increase the bioavail- The dilution and analyzing methods are the same as above.
ability and decrease systematic toxicity of chemotherapeutic drugs for
HCC treatment and possibly for other cancers including PDAC. More-
over, beyond drug delivery capacity, lipiodol also have the potential to 2.3. Drug release test
function as radiosensitizers owing to its high Z element iodine compo-
sition. Our prior research revealed that when lipiodol is emulsified with Firstly, standard curve of gemcitabine based on UV–vis absorption
aqueous solution, it showed a significant radiosensitizing effect on spectrometer (HITACHI) is determined. Gemcitabine hydrochloride was
different types of tumors by producing extra ROS and inducing accurately weighed and dissolved in physiological saline, and accurately
enhanced DNA damage. [15] Accordingly, lipiodol may hold promises diluted to concentrations of 30, 25, 20, 15, 10, and 5 μg/mL as standard
to function as a sensitizer for both radiotherapy and chemotherapy. solution. The absorbance was measured at 268 nm for three times of
Here, we for the first time repurpose the old drug lipiodol as an each concentration, and the average number was used to draw the
efficient chemoradiosensitizer in the treatment of PDAC. Gemcitabine, a standard curve.
drug that blocks DNA replication and repair, is chosen as the drug The in vitro release curve of gemcitabine loaded IOE was studied by a
representative as it is a first-line treatment with radiotherapy for pa- dialysis method. In short, 1 mL IOD is put into a dialysis bag (MWCO =
tients with pancreatic cancer. [16,17] To increase the solubility and 10,000 Da) and immersed into 200 mL physiological sodium. For pH 5.5
stability of water-soluble drug in lipiodol, a stable emulsion using clin- solution, HCl (13 M) was used to adjust pH. Sample is placed on a stir-
ically recommended emulsifier is designed. Such a constructure fulfills ring platform at room temperature with constant magnetic stirring (400
all the criteria for enhancing chemo- and radiotherapy of pancreatic rpm). 5–20 mL of sodium solution is withdrawn at different time points
cancer. Firstly, the iodine-rich lipiodol emulsion showed an improved and analyzed by UV spectrometer at λ = 268 nm. Fresh sodium media
radiotherapeutic effect in comparison with the clinically approved HfO2 was supplemented to an equal volume.
nanosensitizer. Secondly, it functions effectively as a drug carrier,
causing more substantial DNA damage than the administration of free
drug alone in in vivo study. Moreover, the homogenous spatial distri- 2.4. In vitro free radical detection
bution and long-term retention ability were also observed in solid PDAC
tumors, which is beneficial for improving drug bioavailability and long The ROS probe 2 ‘, 7’ - dichlorodihydrofluorescein diacetate (DCFH-
period fractionated radiotherapy in practical treatment of pancreatic DA, 10 mM) was used to evaluate the ability of IOE to enhance the
cancer. These features enable lipiodol emulsion to realize significant generation of free radicals after irradiation in vitro. Firstly, DCFH-DA
tumor inhibition and enhanced survival rate of PDAC-bearing model was mixed in dimethyl sulfoxide (0.25 mL) and 0.01 M sodium hy-
under concurrent chemo-radiotherapy. At the in vitro level, a ROS- and droxide (1 mL) under dark conditions. After 30 min of reaction at room
DNA damage-related therapeutic mechanisms were studied and verified temperature, the reaction was terminated with phosphate buffer (PBS, 5
using RNA Sequencing analysis. Lastly, a comprehensive toxicity mL) to obtain the working solution. The experimental groups include
assessment of the lipiodol emulsion on survival, body weight, blood control, IOE, X-ray, IOE + X-ray, and the concentration of IOE is 480 μg/
index and healthy organ including lung and pancreas is conducted. Our mL (dissolved in PBS), X-ray irradiation setting is 6 Gy, 160 kV, 15 mA.
study offers a clinically feasible sensitizer that can simultaneously in- The fluorescence spectrum of DCF was detected using a fluorescence
crease the chemotherapy and radiotherapy of pancreatic cancer. Given spectrophotometer (Horiba, λEx = 488 nm, λEm = 525 nm).

243
S. Zhu et al. Journal of Controlled Release 374 (2024) 242–253

2.5. Animal Feeding and ethical regulations a small animal CT imaging system (PerkinElmer). HU value was ob-
tained. In vivo CT imaging was performed using Panc02 tumor bearing
C57BL/6 (5–6 weeks) mice were purchased from Beijing HFK mice with intratumoral injection of 10 μL IOE. CT imaging (Triumph X-
Biotechnology Co., Ltd. The feeding conditions: temperature 25 ± 2 ◦ C, SPECT/X-O CT) was performed on day 0 (1 h after injection) and 10 days
normal light cycle (12 h light/12 h darkness), raised in polypropylene respectively. The study of intratumoral distribution of radiosensitizers is
cages equipped with sterilized feed and mineral water. All animal ex- based on synchrotron radiation μCT. Same volume (10 μL) of HfO2 or
periments were conducted in accordance with the guidelines issued by IOE was injected into the Panc02-bearing tumor mice. Then the tumor
the Key Laboratory for Biomedical Effects and Nanosecurity of Nano- was obtained after 2 h and stored at − 80 ◦ C. Synchrotron radiation X-ray
materials, Chinese Academy of Sciences (Institute of High Energy microscopy imaging device (Beijing Synchrotron Radiation Facility,
Physics, Chinese Academy of Sciences), and the animal ethics approval 4W1A line station) was used for μCT imaging, data was analyzed on
number was IHEPLLSC-033. The tumor size shall not exceed the Avizo 8.0 software.
maximum volume of 1500 mm3.
2.11. Cell culture
2.6. Tumor volume monitoring
All cells (including cancer cells and healthy tissue cells) were ob-
Body weight and tumor volume were monitored in all tumor treat-
tained from the Cell Resource Center, Institute of Basic Medical Sciences
ment experiments every two or three days. The tumor volume (V) was
and cultured in DMEM medium containing 10% fetal bovine serum
measured with a vernier caliper and calculated as V = (Width2 ×
(FBS) and 1% penicillin-streptomycin antibody solution at an ambient
Length)/2.
temperature of 37 ◦ C (a humid environment containing 5% CO2).
2.7. Radiosensitizing assessment at in vivo level
2.12. Cell viability measurement
Panc02 cells (5 * 106) were subcutaneously inoculated on the left
abdomen of C57 mice. When the tumor volume reached 60–100 mm3, Cytotoxicity of gemcitabine hydrochloride: Panc02 cells were plated
the mice were randomly divided into three groups (n ≥ 6): (i) saline + X- into a 96 well plate (6 × 103 cells/well). After 24 h, different concen-
ray, (ii) HfO2 + X-ray, (iii) IOE + X-ray. Subsequently, 25 μL different trations of gemcitabine hydrochloride (0, 25 nM, 50 nM, 100 nM, and 1
solutions (saline, HfO2 or IOE) were injected into the tumor and fol- μM) were added to replace the original culture medium (n = 6). After
lowed with X-ray irradiation (4 Gy, 160 kV, 15 mA) for three consecu- incubation for 24 h, the cells were rinsed with PBS, 10% of CCK-8 me-
tive times (once every two days). One or two mice were randomly dium were added and incubate for 1.5–2 h. The absorbance of the su-
selected from each group two days after radiotherapy treatment for lung pernatant at 450 nm was measured with a microplate reader (n ≥ 5,
and tumor analysis. The tissues were fixed in 4% paraformaldehyde for Thermo Scientific, Multiskan MK3).
tissue section. The hafnium oxide nanoradiosensitizer was obtained Cytotoxicity of IOE: HUVEC (Human Umbilical Vein Endothelial
from previous study [15]. Cells) cells were planted into 96 well plates (6 × 103 cells/well). After
24 h, IOE with different concentrations (0, 30, 60, 120, 240, 480 and
2.8. Chemosensitizing assessment at in vivo level 960 μg/mL) were added (n = 6). After another 24 h, wells were rinsed
with PBS and serum-free medium with 10% CCK-8 were added. The cells
Panc02 cells (5 * 106) were subcutaneously inoculated on the left were incubated for 1.5–2 h and measure the absorbance at 450 nm using
abdomen of C57 mice. When the tumor volume reached 50–100 mm3, a microplate reader.
the mice were randomly divided into three groups (n ≥ 6): (i) saline, (ii)
pure drug (gemcitabine hydrochloride, 35 mg/mL), (iii) IOD. 25 μL of
2.13. IOE/IOD preparation for cell use
different treatment was injected intratumorally. The concentration and
dosage of gemcitabine hydrochloride are determined based on drug
The IOD used at cellular level contain different gemcitabine con-
solubility and clinical reference (In clinical study, 2.5 mL of 38 mg/mL
centration from that in vivo. According to the cytotoxicity measurement
gemcitabine was directly injected to pancreatic tumor site on patient.
results of gemcitabine at cellular level, the concentration of 100 nM can
1% amount of clinical dose was used on mice [19]. On the second day
cause effective Panc02 cell killing (70% survival rate). In order to ensure
after drug treatment, one or two tumors were randomly selected from
IOD can simultaneously exert the sensitizing effects of both radiotherapy
each group and fixed in 4% paraformaldehyde for tissue analysis.
and chemotherapy at cell level, the concentration of lipiodol is selected
as 120 μg/mL, with a gemcitabine drug concentration of 100 nM
2.9. Concurrent chemoradiosensitizing assessment
(Fig. S7). Therefore, when the lipiodol concentration in IOD is 271.7
mg/mL, the concentration of gemcitabine should be 226.42 μM. The
Panc02 cells (5 * 106) were subcutaneously inoculated on the left
preparation method of IOD is the same as mentioned above. Briefly, 5.4
abdomen of C57 mice. When the tumor volume reached 50–100 mm3,
mL lipiodol, 3.6 mL of saline containing 179.8 μg/mL gemcitabine, 108
the mice were randomly divided into six groups (n ≥ 10): (i) saline, (ii)
μL Tween 80 and 432 of μL HCO-40 were mixed and sonicated for 5 min
IOD, (iii) saline + X-ray, (iv) pure drug (35 mg/mL) + X-ray, (v) IOE +
and homogenized for another 5 min.
X-ray, and (vi) IOD + X-ray. 25 μL of different solutions were injected
intratumorally, and the mice (only tumor area) were exposed to X-ray
irradiation (8 Gy, 160 kV, 15 mA). On the second day after treatment, 2.14. Cloning experiment
1–2 tumors were randomly selected from each group and fixed in 4%
paraformaldehyde for tissue slice analysis. Tumor volume was moni- Panc02 cells were planted in a 6-well plate (1 × 103 cells/well) and
tored until euthanasia of the mice (tumor volume reached 1500 mm3) or incubated for 24 h. Cells were added with different treatment: control,
60 days after tumor inoculation. pure drug (100 nM), IOE (lipiodol concentration 120 μg/mL), IOD, X-
ray, pure drug + X-ray, IOE + X-ray, IOD + X-ray. Cells were exposed to
2.10. CT and μCT imaging 4 Gy X-ray irradiation (160 kV, 15 mA) after 6 h. After 10 days, cells
were washed with PBS and fixed by 4% paraformaldehyde for 10 min.
For in vitro CT imaging, 1 mL of different solutions, including water, After washing twice with PBS, cells were incubated with Giemsa staining
HfO2, and IOE were put into a 1.5 mL centrifuge tube and imaged under solution for 20 min, with three replicates in each group.

244
S. Zhu et al. Journal of Controlled Release 374 (2024) 242–253

2.15. Cellular uptake experiment Co. (Guangzhou, China).

IOE containing Nile Red dye was used to study the cellular uptake 2.19. Western blot
experiment. Nile Red (MedChemExpress) was mixed with DMSO as
storage solution (1 mM). Herein, to prepare IOE, lipiodol was firstly Panc02 cells were treated with either control or IOD (100 nM gem-
mixed with 135 μM Nile Red storage solution before mixing with saline citabine) for 36 h and then treated with radio immunoprecipitation
and surfactant, as well as mechanical sonification and homogenization. assay lysis buffer on ice. Protein concentration was determined by
For control group, Nile Red storage solution (135 μM) was mixed with Bradford method. Electrophoresis was conducted in 12% SDS poly-
saline (9 mL) and surfactants, and perform sonification and homogeni- acrylamide gel electrophoresis gels at 120 V for 90 min. Proteins are
zation using the same method as IOE. Then Panc02 cells were planted then transferred to polyvinylidene difluoride membranes and blocked
into confocal culture dish (2 × 105 cells/well). After 24 h, different with 5% skim milk for one hour. The membranes were incubated with
solutions (500 nM Nile Red containing control and IOE) were added and primary antibodies of β-actin (Beyotime), Ras (#67648, Cell Signaling
incubated for 6 h. Then Hoechst 33342 (1:100) was added to each group, Technology) and Stat1 (#9172, Cell Signaling Technology) for 12 h at
incubate for another 20 min. Lastly, cells were washed with PBS three 4 ◦ C. Afterwards, membranes were incubated with goat anti-rabbit IgG
times and observed under a confocal microscope. secondary antibody for an hour. After wash, immunopositive blot were
visualized using chemiluminescence detection method.
2.16. Free radical detection
2.20. Long-term safety evaluation
Panc02 cells were planted into confocal culture dishes (2 × 105 cells/
well). 24 h later, the cells were divided into eight groups: control, pure In order to study the long-term in vivo biological safety of IOE,
drug (100 nM), IOE (lipiodol concentration 120 μg/mL), IOD, X-ray, healthy C57 mice were used. 25 μL saline, lipiodol or IOE were intra-
pure drug + X-ray, IOE + X-ray, IOD + X-ray. After incubate with venously injected (n = 6). Body weight was recorded every three days
different solution for 6 h, cells were washed twice with PBS, and 1 mL of and animal behavior were observed every day. Mice was sacrificed on
serum-free medium containing DCFH-DA (1:1000) and Hoechst 33342 the 15th day and whole blood was collected for subsequent blood
(1:100) were added to each group. After incubating for 20 min, the cells routine and biochemical analysis. Major organ including heart, liver,
were rinsed three times with serum-free DMEM, and then irradiated spleen, lungs, and kidneys were collected and fixed in 4% para-
with X-ray (4 Gy, 160 kV, 15 mA). Finally, intracellular ROS generation formaldehyde for histopathological section analysis.
was detected using confocal microscopy (A1/LSM-Kit, Nikon, Japan/ For the study of IOE on the long-term safety of pancreas, after
Germany) and fluorescence images were captured. anesthesia, abdomen of the mouse was opened to expose pancreas, and
IOE or saline (25 μL) were injected directly into the pancreas. Subse-
2.17. DNA damage quently, the abdominal wound of the mouse was surgically sutured.
After 15 days, the mice were euthanized, and pancreas was removed and
Panc02 cells (1 × 105 cells/well) were planted onto a 24 well plate fixed in 4% paraformaldehyde for histopathological section analysis.
covered with glass coated with L-Polylysine (L-PLL). 24 h later, the cells
were divided into eight groups: control, pure drug (100 nM), IOE (lip- 2.21. Data statistical analysis
iodol concentration 120 μg/mL), IOD, X-ray, pure drug + X-ray, IOE +
X-ray, IOD + X-ray. After incubating with different solution for 6 h, the All data are expressed as mean ± standard deviation. Three or more
cells are exposed to X-ray irradiation (4 Gy, 15 mA, 160 kV). After 1 h of groups were compared using one-way analysis of variance (ANOVA) and
cultivation, cells were fixed with 4% paraformaldehyde for 10 min, Turkey’s post hoc test. The comparison between the two groups was
0.5% Triton X100 for 10 min, 5% FBS (diluted with PBS) for 1 h. Then conducted using Student’s t-test. Unless otherwise specified, asterisks
the cells were incubated with phospho-histone H2A.X (Ser139) rabbit indicate significance: * p < 0.05, * * p < 0.01, * * * p < 0.001, * * * p <
mAb (1:800) at 4 ◦ C for 12 h (shaking slowly), followed by Cy3 labeled 0.0001.
goat anti-rabbit IgG (H + L) (1:500) incubation at room temperature for
1 h. Finally, Hoechst 33342 (1:200) was added and incubated at room 3. Result and discussion
temperature for 10 min. After sealing, fluorescence images were
captured using confocal microscopy (A1/LSM-Kit, Nikon, Japan/ 3.1. Characterization of lipiodol emulsion
Germany).
As elaborated above, PDAC is featured with a high degree of fibrosis
2.18. RNA Seq analysis and abnormal vascular structure, which makes it difficult for drugs to
effectively penetrate into the tumor tissue. To solve this problem, the
Panc02 cells (1 × 105 cells/well) were planted onto a 6 well plate. intratumor administration is an effective way to ensure the high con-
After 24 h, the cells were treated with either control or IOD (100 nM centration of drug in tumor. In addition, when designing drug carrier
gemcitabine) for 36 h. Then total RNA was extracted using Trizol re- emulsion, it is necessary to consider how to improve the drug retention
agent kit (Beyotime) according to the manufacturer’s protocol. RNA efficiency and enhance the drug diffusion in the tumor. Consequently,
quality was assessed on an Agilent 2100 Bioanalyzer (Agilent Technol- this work aims to design a lipiodol emulsion (IOE) with small size and
ogies, Palo Alto, CA, USA) and checked using RNase free agarose gel low fluidity. The small size is conducive to better passing through the
electrophoresis. After total RNA was extracted, eukaryotic mRNA was fibrotic tumor stroma and improving the drug diffusion in tumor tissue,
enriched by Oligo(dT) beads. Then the enriched mRNA was fragmented while appropriate viscosity can help to stay at the tumor site after in-
into short fragments using fragmentation buffer and reversely tran- jection, extending the retention time and concentration of drugs in the
scribed into cDNA by using NEBNext Ultra RNA Library Prep Kit for tumor, and improving the retention efficiency of drugs. According to
Illumina (NEB #7530, New England Biolabs, Ipswich, MA, USA).The previous research, the surfactant involved in the formulation of lipiodol
purified double-stranded cDNA fragments were end repaired, A base emulsion should have a relatively high Hydrophilic Lipophilic Balance
added, and ligated to Illumina sequencing adapters. The ligation reac- (HLB) and hydrophilicity to increase the dispersibility of IOE in aqueous
tion was purified with the AMPure XP Beads(1.0×). And polymerase media. [20] For lipid phase surfactant, the density is preferably over
chain reaction (PCR) amplified. The resulting cDNA library was 0.95. [20] Therefore, in this study, Tween 80 (HLB = 15) and polyoxyl
sequenced using Illumina Novaseq6000 by Gene Denovo Biotechnology 40 hydrogenated castor oil (HCO 40, density 0.96) were chosen as the

245
S. Zhu et al. Journal of Controlled Release 374 (2024) 242–253

emulsifiers. [21] Both are approved for clinical application, indicating ray itself, indicating its potential for enhancing radiotherapeutic effect.
good safety. [22,23] Next, the average particle size, polydispersity (PDI) Moreover, in order to measure the drug release behavior of drug-loaded
and zeta potential of IOE in water were characterized by dynamic light IOE (IOD), UV standard curve of gemcitabine was first created (Fig. S3).
scattering (DLS). As shown in Fig. 1a, the average particle size of IOE is The standard curve shows a good linear relationship. Subsequently, drug
~400 nm (zeta potential is − 58 mV, PDI is 0.2) and remains unchanged release ability of IOD was measured using a dialysis method, which is
over 10 days of storage (Fig. S1), indicating its good stability. Such one of the most commonly used methods to study drug release behavior.
nanosized property is also considered beneficial for efficient tumor [30]. From the cumulative drug release curve shown in Fig. 1f, gemci-
cellular uptake. [24] In addition, needle injection during in vivo appli- tabine from IOD exhibited a stable and sustained release ability, where
cation can cause shear stress and pressure changes in emulsion, which drug release rate can reach 50% and 90% at day 4 in normal and acidic
may affect their physicochemical properties. [25] We then studied the environment, respectively. Such slow and sustained drug release char-
size distribution property of as-prepared emulsion after passing through acteristics of IOD offer an opportunity to realize prolonged interaction
the syringe needle, where the average size showed a similar pattern between drug and tumor, so as to increase the therapeutic effect of
(Fig. 1b) to before needle injection (Fig. 1a). The stability of the nano- chemotherapy. The above characterizations indicate that the stable IOE
emulsion is critical for clinical efficacy and safety, especially considering may function as both radiosensitizer and chemosensitizer for in vivo
the large lipiodol oil droplet may cause unintentional vascular occlu- application.
sion. [26] Thus increasing stability to avoid coalescence is rather
important in this scenario. Then, we studied the rheology profile of IOE, 3.2. Radio- and chemo-sensization evaluation of lipiodol emulsion
where the formulation exhibits certain viscosity and pseudoplastic flow
with shearing-thin property (Fig. 1c and Fig. S2). This is consistent with Before assessing the radiochemosensitization effect of IOE, we first
the properties of clinically prepared lipiodol emulsion. [27] Such thix- investigated the radiosensitization and chemosensitization potential of
otropical property can not only enhance the stability of emulsion IOE separately in a classical tumor regrowth inhibition assay. Subcu-
(reduce the flow of droplets) [28], but also extend the retention time of taneous Panc02 PDAC tumor-bearing C57BL/6 mouse model was
drug-loaded emulsion at the administration site so as to improve the established. To better illustrate the radiosensitizing ability of IOE, HfO2
bioavailability and therapeutic efficacy of pharmaceutical formulations. nanoradiosensitizer was used for comparison. All groups received a total
[29] After evaluating the emulsion property, we further studied the 12-Gy radiation (3 * 4 Gy, multi-times radiation to mimic clinical frac-
radiosensitizing and drug carrier ability of IOE. We first compared the X- tionated radiotherapy) with or without intratumorally-injected radio-
ray absorption ability of IOE with clinically used HfO2 nano- sensitizers (Fig. 2a). On comparing tumor volume plot, IOE + X-ray
radiosensitizers. Owing to the high concentration of iodine (271 mg/ treatment was considerably more effective than X-ray and HfO2 + X-ray
mL), IOE exhibited a much stronger CT signal than that of HfO2 solution treatment (P = 0.01) in inhibiting tumor growth. The H&E staining of
(53.3 mg/mL) (Fig. 1d). This is one of the major advantages of IOE as a tumor slices also indicated the most significantly damaged architecture
liquid radiosensitizer since such a high working concentration is not and cell death in the IOE + X-ray radiation group (Fig. 2b). Moreover, at
possible to be realized by other particle-based radiosensitizers due to its the end of the experiment, ascites (extra fluid built up in the abdomen)
technical limitations. Meanwhile, given the radiosensitizing mechanism was found in one of the HfO2 + X-ray treated group (Fig. S4). We then
of high Z-based materials is related to the electron-mediated ROS pro- analyzed the lung metastasis of different groups. As shown in Fig. 2b,
duction, we measured the ROS-generating ability of IOE under X-ray both X-ray (two out of six) and HfO2 + X-ray (one out of six) treated
irradiation. As shown in Fig. 1e, IOE produced much higher ROS than X- groups showed aggressive lung metastases. In comparison, no obvious

Fig. 1. Characterization of IOE. (a) Hydrodynamic size distribution, ζ potential (unit: mV) and polydispersity (PDI). (b) Hydrodynamic size distribution of IOE after
needle injection. (c) viscosity-shear rate curves. (d) CT values in Hounsfield units (HU) of the indicated solutions. (e) Total radical oxygen species measurement by
DCFH-DA probes with different treatments. (f) Drug release profile under different pH environment. IOE: lipiodol emulsion.

246
S. Zhu et al. Journal of Controlled Release 374 (2024) 242–253

Fig. 2. Radio- and chemo-sensitization evaluation of lipiodol emulsion (a) Average tumor volumes of Panc02 xenograft tumors for radiosensitization assessment. (b)
Representative images of hematoxylin and eosin (H&E, scale bars 100 μm) of Panc02 tumor and lung organ after different treatments. Average tumor volumes (c) and
whole slide immunohistochemical γ-H2AX staining (d) of Panc02 xenograft tumors after chemotherapy (scale bar: 2000 μm (left) and 1000 μm (right)). (e) In vivo CT
images taken at different days after intratumoral injection of IOE. (f) Intratumoral distribution analysis by synchrotron radiation-based micro-CT of the indicated
treatments. IOE: lipiodol emulsion; IOD: drug-loaded lipiodol emulsion.

pulmonary metastasis was observed in the IOE + X-ray group, further signals. IOE-assisted drug delivery realized a more significant and
demonstrating the enhanced radiotherapeutic effect of IOE. This result is extensive chemotherapeutic effect (DNA damage) than free drug itself.
in consistent with the X-ray absorption ability (Fig. 1d), where highly This may be due to the fact that free drugs can easily diffuse or be
concentrated iodine in IOE could offer greater interaction with incident flushed away by local arterial blood flow [34], leading to limited
photons than HfO2 at a given volume in tumors. Next, we evaluated the treatment of tumors. In contrast, using small-sized IOE as a carrier can
chemosensitizating ability of IOE by comparing the therapeutic effect of realize slow drug release (Fig. 1f), and enhance the retention and
IOD and free drug on Panc02 tumors (Fig. 2c). Although two drug- diffusion of therapeutic agents at tumor site, thereby increasing drug
treated groups showed similar trends in tumor inhibition outcome, bioavailability and maximizing the cytotoxic damage to cancer cells.
histological analysis revealed the improved therapeutic effect of IOD. As In order to confirm the above demonstration, we further studied the
the major mechanism of action of gemcitabine is the inhibition of DNA retentivity and diffusion of IOE at tumor site. Radiopaque IOE in solid
synthesis [31], γ-H2AX which can represent DNA damage level is tumor was well visualized under CT scanning and was retained within
considered as a marker for gemcitabine-induced toxicity. [32] The the tumor for over ten days with negligible intratumoral redistribution
expression of γ-H2AX protein could indicate the induction of DNA (Fig. 2e). No redistributing or accumulating in surrounding non-
double-strand breaks. [33] We thus evaluated the γ-H2AX expression of malignant tissues was observed. This property is rather important not
different groups. From the fluorescent imaging of the full tumor scan only in chemosensitization but also in radiosensitization, as it can ensure
(Fig. 2d), the IOD group exhibited a large area of obvious γ-H2AX the radiosensitizing effect in clinical practice. The prolonged retention
expression, while the free drug group only showed scattered fluorescent ability could provide a wide therapeutic window for fractionated

247
S. Zhu et al. Journal of Controlled Release 374 (2024) 242–253

radiotherapy. [35] Moreover, from Fig. 2d, a large portion of tumor 3.3. Combined chemoradiotherapy evaluation of lipiodol emulsion
exhibited DNA damage signal in IOE-high group, indicating a wide-
spread distribution of drug in solid tumor. We then examined the tumor The above result drove use to further examine the simultaneous
distribution ability of IOE. It is believed that the degree of high Z sensitizing ability of IOD in chemoradiotherapy. Tumor-bearing mice
elements-based radiation sensitization also greatly depends on the were randomly assigned into six groups: control, IOD, X-ray, D (pure
spatial distribution of the elements within the tumor [36]. Therefore we drug) + X-ray, IOE + X-ray and IOD + X-ray. Mice in the X-ray groups
compared this property of IOE with HfO2 nanoradiosensitizers. As is received 8 Gy radiation 2 h after a single intratumoral administration of
presented in Fig. 2f, IOE demonstrated a wider and more homogenous different treatments. Tumor volumes were measured routinely to assess
spatial distribution than HfO2 nanoparticles. This may be due to the treatment effectiveness. As shown in Fig. 3a, all X-ray-based groups
liquid nature of IOE in comparison to solid particles, where the inherent showed obvious tumor inhibition than the control and IOD group,
fluidity and mobility of liquid-natured agents allow for a more uniform showing partial treatment efficacy. On comparing the curves of different
dispersion throughout the tumor mass. Such ability could ensure more treatments with X-ray, the addition of IOE resulted in far smaller tumor
cancer cells to be exposed to the therapeutic treatment (radiation or volumes than X-ray alone, in consistent with the previous radio-
chemo-drug), leading to a more effective and consistent response to the sensitizing result (Fig. 2a). Free drug administration also slowed tumor
treatment. Taken together, our data suggests that IOE could function as regrowth, but the effect is far less sufficient than IOD treatment. As
an effective sensitizer for enhancing the radiotherapy and chemotherapy expected, IOD + X-ray group witnessed the greatest delay on tumor
of pancreatic cancer respectively. The homogenous accumulation and growth, indicating the notable therapeutic effect of IOE-mediated
prolonged retention ability at tumor site are the key factors for combined chemoradiotherapy (P = 0.01 with X-ray; P = 0.03 with D
enhancing the in vivo efficacy of IOE. + X-ray). Remarkably, three of eight tumors treated by IOD + X-ray

Fig. 3. Chemoradiosensitization of lipiodol emulsion at in vivo level. Average tumor volumes (a), individual tumor volumes (b) and survival rate curve (c) of Panc02
xenograft tumors with the indicated treatments at different time points (days). (d) Representative images of hematoxylin and eosin (H&E, scale bars 30 μm) and
immunohistochemical staining (Tunel, Ki67, TUNEL, and γ-H2AX, scale bars 50 μm) of Panc02 xenograft tumors with the indicated treatments. IOE: lipiodol
emulsion; IOD: drug-loaded lipiodol emulsion.

248
S. Zhu et al. Journal of Controlled Release 374 (2024) 242–253

were eradicated within four weeks (Fig. 3b). At the end of the experi- healthy cell line HUVEC at a concentration as high as 960 μg/mL
ment, all tumor-bearing mice were died expect IOE/IOD-based radio- (Fig. 4b). These results indicated the safety and suitability of IOE for in
therapy treatment (Fig. 3c). IOE + X-ray group resulted in a 22% vitro study. For assessing the radiosensitization ability of internalized
survival rate while half (50%) of the mice in IOD + X-ray group survived IOE, a classic colony formation assay was employed on Panc02 cells. As
from PDAC tumor inoculation. This further validates the therapeutic displayed in Fig. 4c and Fig. S8, the treatment with IOE + X significantly
effect of IOE-sensitized radiotherapy and the synergistic effect of com- reduced the number of colonies when compared to X-ray group (p <
bined treatment based on IOD-mediated radiochemotherapy. In addi- 0.0001), proving its effective radiosensitivity. IOD + X caused the least
tion to tumor volume measurement, we also performed histological colony formation, further proving that cell proliferation ability was
analysis to confirm the therapeutic effects. Tumors from different groups greatly compromised by the combined therapy. Herein, at cellular level,
were subjected to Hematoxylin-eosin (H&E), terminal deoxynucleotidyl pure drug and IOD with the same gemcitabine concentration exhibit
transferase dUTP nick end labeling (TUNEL), Ki67 and γ-H2AX staining, similar killing effects. There are two reasons for this phenomenon.
which allows studying of the tumor tissue’s general characteristics, cell Firstly, unlike in vivo experiments which can be influenced by blood
necrosis, proliferation and DNA damage profile, respectively (Fig. 3d flush, free drug has the same accumulation and distribution behavior as
and Fig. S6). As expected, the group receiving IOD + X-ray irradiation IOD at in vivo level. Secondly, in cell experiments, lipiodol emulsion was
caused the most severe cancer necrosis from H&E result. Both IOE-based added to culture medium using vigorous pipetting to form even distri-
radiotherapy groups showed considerably more TUNEL-positive cells bution, which may destroy the emulsion stability and promote drug
than other groups, while IOD + X-ray treatment resulted in the most release, leading to similar killing effect to pure drug at cellular level.
significant apoptotic activity. On the other hand, Ki67 staining result Next, ROS production and DNA damage levels were also evaluated at the
revealed the great suppression of tumor cell proliferation ability in IOE/ Panc02 cell (Fig. 4d). DCFH-DA probe was used to measure intracellular
IOD + X-ray-treated tumors, which is consistent with the observation on ROS production, which produces green fluorescence when exposed to
tumor growth. Lastly, we studied the DNA damage profile using γ-H2AX ROS. [37] As expected, IOE significantly improved the X-ray-induced
as DNA is the therapeutic target of both gemcitabine and high Z-based ROS production, while the strongest fluorescent signals were observed
radiosensitizer. Similar to the result in Fig. 2d, IOD caused more sig- in IOD + X-ray treatment (Fig. S9a). Under the given drug (100 nM) and
nificant γ-H2AX expression than the free drug. IOE + X-ray group also X-ray (4 Gy) dose, gemcitabine showed a limited role in ROS generation
showed more DNA toxicity than X-ray itself. These results further vali- ability, but this is not the case when it comes to DNA damage. γ-H2AX
date the chemoradiosensitizing ability of IOD from a histological foci results indicate that gemcitabine could cause significant DNA
perspective. double-strand breaks even without X-ray radiation ((Fig. S9b). More-
over, IOE significantly increased X-ray caused DNA damage, and IOD
resulted in the most evident DNA damage, confirming the previous
3.4. Act of mechanism study at cellular level result. As apoptosis is the major cell death modality in response to DNA
damage, we also studied the apoptosis-induction ability of IOD under X-
Next, we studied the IOE-based radiochemosensitization at cellular ray. As shown in Fig. S10, both IOE and IOD could induce more signif-
level to confirm the above result and explore the corresponding under- icant cell apoptosis (green fluorescence) under X-ray radiation than
lying mechanisms. We firstly evaluated the tumor uptake and cellular radiotherapy alone, while IOD could exacerbate the damaging effect to
biocompatibility of IOE. As shown in Fig. 4a, the internalization of IOE late apoptosis and necrosis (green/red double fluorescence). The above
was obviously observed in Panc02 cell after 6 h incubation, indicating its experiments indicate that IOD functions to improve ROS generation,
efficient cellular uptake. It also showed good biocompatibility on

Fig. 4. (a) Cellular uptake of IOE by Panc02 cells. (b) Cell viability of IOE on HUVEC. (c) Clonogenic survival result of different treatments on Panc02 cells. (d) DCF
fluorescence images and γ-H2AX foci of Panc02 cells after different treatments (scale bar 25 μm). IOE: lipiodol emulsion; D: free drug; IOD: drug-loaded lip-
iodol emulsion.

249
S. Zhu et al. Journal of Controlled Release 374 (2024) 242–253

DNA damage and apoptosis so as to inhibit tumor cell growth. These upregulated after treatment. Protein expression of representative genes
results provide validation for the efficacy of in vivo chemoradiotherapy. (Ras and Stat1) was also verified using Western blot (Fig. S12). These
To further investigate the molecular mechanism of IOD-mediated genes are also reported to be associated with radiation and gemcitabine
radiochemosensitizating effect, we performed RNA-Seq to elucidate treatment in previous studies [39–43]. The RNA-seq results provide
the critical molecular pathways. The distribution of differentially molecular validation for the functioning mechanism of IOD-mediated
expressed genes (DEGs) between control and IOD + X-ray treated groups chemoradiotherapy, where IOD + X-ray could induce cell death by
are presented as heat maps and volcano plots. As shown in the clustering increasing ROS production and causing DNA damage in Panc02 cell.
heatmap (Fig. 5a), the expression level of the homeologs in triple sam-
ples exhibited a consistent trend. Volcano plots indicated 666 up-
regulated genes and 72 down-regulated genes among DEGs (Fig. 5b 3.5. Safety evaluation of lipiodol emulsion
and Fig. S11). To better illustrate the molecular mechanism of IOD + X-
ray treatment, different enrichment analysis of DEGs were utilized. Gene Desirable biosafety is one of the most important factors of newly
Ontology (GO) annotation analysis revealed that IOD-mediated che- developed medicine for clinical translation. For lipiodol-based emulsion
moradiotherapy could cause the response of Panc02 cell to stress, in pancreatic cancer treatment, two major concerns are systematic and
external stimulus, chemical stimulus, and immune (Fig. 5c). In addition, pancreatic toxicity. Based on clinical practice, pulmonary embolism may
the most significantly changed biological pathways were analyzed based occur from inadvertent systemic vascular injection of lipiodol, which
on Kyoto Encyclopedia of Genes and Genomes (KEGG) database. From may cause pulmonary infarction or even fatalities. [44] Therefore,
Fig. 5d, the overall changes in cellular and organism-level function be- reducing the systematic toxicity of lipiodol-based agents is a key priority
tween different groups lie in the immune system, signal transduction, in providing safe therapy. Considering the small size and good stability/
metabolism, cancer (overview) and cell growth and death. [38] In terms dispersibility of IOE in aqueous solution, we suppose it may have the
of cellular signaling pathways, DEGs were closely associated with ROS ability to avoid lipiodol-associated adverse effects. We then compared
oxidative stress (MAPK, PI3K-Akt, NF-kappa B and FoxO signaling its safety profile with lipiodol. As seen in Fig. 6a, after intravenous in-
pathway), DNA damage (JAK-STAT signaling, cytosolic DNA sensing jection of different treatment, lipiodol caused 40% sudden death within
and P53 signaling pathways), and immune-response or apoptosis/ne- 2 days and obvious body weight change was still witnessed at day 4
crosis pathways. Lastly, we studied the expression of the key genes as (Fig. S13). In comparison, mice treated with IOE showed no sudden
well as corresponding up/down-stream genes to confirm the above death and less body weight. After two weeks of observation, blood
signaling pathways. As shown in Fig. 5f, DNA damage-related genes samples and major organs are obtained for analysis. From Fig. 6b,
(IFNb, Stat1 and 2, Ddit3, Rad52 and Nr1d1), stress response-associated obvious infiltration was observed in lipiodol group, indicating an in-
genes (Nfkb, Ppp1r15a, PLK, Foxo3, EGR1, Txnip and Ras), and flammatory lung injury. As expected, IOE injection did not induce
apoptosis/necrosis-related genes (Bcl2, Fas, Trp53inp1 and Tnfrsf26) appreciable toxicity to the lung. Moreover, no significant difference
exhibited significant alteration between control and IOD + X-ray group, between mice treated with control and IOE was found across all tested
indicating the increase DNA damage, stress response and apoptosis are biochemical and hematological parameters (Fig. 6c and d). Histopath-
ological assessment of normal organs including heart, liver, spleen, and

Fig. 5. RNA Sequence analysis of different treatments on Panc02 cell. Venn map (a) and Volcano plot (b) of the DEGs of different groups. (c) The most 10
significantly changed GO terms in the GO enrichment analyses. Overall changes in cellular and organism-level (d) and top 12 cell signaling pathways (e) based on
KEGG enrichment analyses. (f) The differentially expressed mRNAs in different group. Each row represents a single mRNA, and each column represents one tissue
sample. Red denotes high relative expression and blue indicates low relative expression. Ctrl: control, ID: IOD + X-ray treatment. (For interpretation of the references
to colour in this figure legend, the reader is referred to the web version of this article.)

250
S. Zhu et al. Journal of Controlled Release 374 (2024) 242–253

Fig. 6. Long-term in vivo toxicologic assessment of lipiodol emulsion. (a) Survival rate curve during experiment period (n = 6). (b) Representative histopathological
images of lung organ after treatment. Blood biochemistry (c) and hematological parameters (d) of different treatments. (e) Representative histopathological images of
heart, kidney, liver, spleen (scale bar 100 μm), and pancreas sections (scale bar 70 μm). C: control; IO: lipiodol; IOE: lipiodol emulsion.

kidney, also indicated no noticeable inflammatory or necrotizing lesions hydrophilic and hydrophobic drug, such as the active metabolite SN38
of IOE treatment (Fig. 6e). The emulsifying method of lipiodol presented of irinotecan commonly used in PDAC treatment. Future research will
in our study may offer a solution to avoid adverse fatal toxicity that is focus on exploring diverse applications of IOE as a multi-drug carrier in
associated with lipiodol in clinics. In addition to systematic toxicity, different cancer treatments.
another primary concern in the treatment of pancreatic cancer is the Supplementary data to this article can be found online at https://doi.
potential toxicity to the healthy pancreatic tissue. [45] Consequently, org/10.1016/j.jconrel.2024.08.020.
pancreas toxicity of IOE was also assessed by directly injecting IOE into
the pancreas. Pancreas was harvested after 14 days and no obvious CRediT authorship contribution statement
damage or abnormalities were found on pancreas from histopathological
analysis (Fig. 6e), suggesting its safe application for pancreatic cancer Shuang Zhu: Conceptualization, Methodology, Formal analysis,
treatment. Based on the above result, it can be inferred that IOE may be Investigation, Visualization, Writing – original draft. Chenglu Gu:
safe and well-tolerable for clinical use. Investigation, Methodology, Visualization, Writing – review & editing.
Long Gao: Resources, Investigation, Methodology, Writing – review &
4. Conclusion editing. Shuanglong Du: Investigation. Duiping Feng: Resources.
Zhanjun Gu: Conceptualization, Supervision, Funding acquisition,
In this study, we for the first time repurposed lipiodol as a simulta- Writing – review & editing.
neous radio- and chemo-sensitizer of pancreatic cancer treatment in
combined with gemcitabine. IOE demonstrated improved effectiveness Declaration of competing interest
over existing radiosensitizers due to its highly-concentrated high Z
element and homogenous spatial distribution. Moreover, the long-term None.
retention and sustained release property in solid PDAC tumors enable
IOD to effectively enhance the drug bioavailability. In vivo results Data availability
demonstrated significant tumor inhibition and survival rate under IOD-
mediated concurrent chemoradiotherapy. The in vitro study further Data will be made available on request.
revealed the ROS and DNA damage-related therapeutic mechanism of
lipiodol-based chemoradiosensitizer. By employing clinically-approved Acknowledgement
agents, the as-prepared IOE exhibited good biocompatibility over lip-
iodol itself. This study offers a clinically feasible sensitizer for simulta- This work was supported by National Key Research and Develop-
neous chemoradiotherapy. Given lipiodol’s broad-spectrum drug ment Program of China (2021YFA1201200 and 2020YFA0710702),
delivery capabilities, IOE may also be suitable for delivering various National Natural Science Foundation of China (22375205), Strategic

251
S. Zhu et al. Journal of Controlled Release 374 (2024) 242–253

Priority Research Program of Chinese Academy of Sciences [17] E.P. Balaban, P.B. Mangu, A.A. Khorana, M.A. Shah, S. Mukherjee, C.H. Crane, M.
M. Javle, J.R. Eads, P. Allen, A.H. Ko, A. Engebretson, J.M. Herman, J.H. Strickler,
(XDB36000000), Directional Institutionalized Scientific Research Plat-
A.B. Benson 3rd, S. Urba, N.S. Yee, Locally advanced, Unresectable pancreatic
form relies on Beijing Synchrotron Radiation Facility of Chinese Acad- Cancer: American Society of Clinical Oncology clinical practice guideline, J. Clin.
emy of Sciences, Beijing Natural Science Foundation (2222087), Oncol. 34 (2016) 2654–2668, https://doi.org/10.1200/JCO.2016.67.5561.
Technology Innovation Program of Institute of High Energy Physics [18] D. Nakmode, V. Bhavana, P. Thakor, J. Madan, P.K. Singh, S.B. Singh, J.
M. Rosenholm, K.K. Bansal, N.K. Mehra, Fundamental aspects of lipid-based
(Y9545130U2), Key Research and Development Projects of Shanxi excipients in lipid-based product development, Pharmaceutics 14 (2022), https://
Province(202102130501014); Shanxi Provincial Clinical Research doi.org/10.3390/pharmaceutics14040831.
Center for Interventional Medicine (202204010501004); Natural Sci- [19] M.J. Levy, S.R. Alberts, W.R. Bamlet, P.A. Burch, M.B. Farnell, F.C. Gleeson, M.
G. Haddock, M.L. Kendrick, A.L. Oberg, G.M. Petersen, N. Takahashi, S.T. Chari,
ence Foundation of Shanxi Province (202203021211021). EUS-guided fine-needle injection of gemcitabine for locally advanced and
metastatic pancreatic cancer, Gastrointest. Endosc. 86 (2017) 161–169, https://
References doi.org/10.1016/j.gie.2016.11.014.
[20] C. ROBIC, J.-F. Mayer, Composition for Vectorizing an Anti-Cancer Agent.
WO2015118113A1, 2015.
[1] R. L. Siegel, K. D. Miller, N. S. Wagle, A. Jemal, Cancer statistics, 2023, CA Cancer
[21] K. Kawakami, T. Yoshikawa, Y. Moroto, E. Kanaoka, K. Takahashi, Y. Nishihara,
J. Clin. Densitom. 73 (2023) 17–48. doi:https://doi.org/10.3322/caac.21763.
K. Masuda, Microemulsion formulation for enhanced absorption of poorly soluble
[2] J.D. Mizrahi, R. Surana, J.W. Valle, R.T. Shroff, Pancreatic cancer, Lancet 395
drugs: I. Prescription design, J. Control. Release 81 (2002) 65–74, https://doi.org/
(2020) 2008–2020, https://doi.org/10.1016/S0140-6736(20)30974-0.
10.1016/S0168-3659(02)00049-4.
[3] A.M. Buckley, N. Lynam-Lennon, H. O’Neill, J. O’Sullivan, Targeting hallmarks of
[22] K. Hippalgaonkar, S. Majumdar, V. Kansara, Injectable lipid emulsions-
cancer to enhance radiosensitivity in gastrointestinal cancers, Nat. Rev.
advancements, opportunities and challenges, AAPS PharmSciTech 11 (2010)
Gastroenterol. Hepatol. 17 (2020) 298–313, https://doi.org/10.1038/s41575-019-
1526–1540, https://doi.org/10.1208/s12249-010-9526-5.
0247-2.
[23] L.S. Schwartzberg, R.M. Navari, Safety of Polysorbate 80 in the oncology setting,
[4] X. Zhou, P. Zhang, N. Liu, X. Zhang, H. Lv, W. Xu, M. Huo, Enhancing
Adv. Ther. 35 (2018) 754–767, https://doi.org/10.1007/s12325-018-0707-z.
chemotherapy for pancreatic cancer through efficient and sustained tumor
[24] E. Sanchez-Lopez, M. Guerra, J. Dias-Ferreira, A. Lopez-Machado, M. Ettcheto,
microenvironment remodeling with a fibroblast-targeted nanosystem, J. Control.
A. Cano, M. Espina, A. Camins, M.L. Garcia, E.B. Souto, Current applications of
Release 361 (2023) 161–177, https://doi.org/10.1016/j.jconrel.2023.07.061.
Nanoemulsions in Cancer therapeutics, Nanomaterials (Basel) 9 (2019) 821,
[5] C. de la Pinta, Stereotactic body radiotherapy in pancreatic adenocarcinoma,
https://doi.org/10.3390/nano9060821.
Hepatobiliary Pancreat. Dis. Int. 23 (2024) 14–19, https://doi.org/10.1016/j.
[25] B. Wahlberg, H. Ghuman, J.R. Liu, M. Modo, Ex vivo biomechanical
hbpd.2023.03.002.
characterization of syringe-needle ejections for intracerebral cell delivery, Sci. Rep.
[6] A.R. Wolfe, T.M. Williams, Altering the response to radiation: radiosensitizers and
8 (2018) 9194, https://doi.org/10.1038/s41598-018-27568-x.
targeted therapies in pancreatic ductal adenocarcinoma: preclinical and emerging
[26] J. Iwazawa, S. Ohue, N. Hashimoto, T. Mitani, Local tumor progression following
clinical evidence, Ann. Pancreat. Cancer 1 (2018), https://doi.org/10.21037/
lipiodol-based targeted chemoembolization of hepatocellular carcinoma: a
apc.2018.08.02.
retrospective comparison of miriplatin and epirubicin, Cancer Manag. Res. 4
[7] J. Peng, B.F. Sun, C.Y. Chen, J.Y. Zhou, Y.S. Chen, H. Chen, L. Liu, D. Huang,
(2012) 113–119, https://doi.org/10.2147/cmar.S30431.
J. Jiang, G.S. Cui, Y. Yang, W. Wang, D. Guo, M. Dai, J. Guo, T. Zhang, Q. Liao,
[27] H. Demachi, O. Matsui, H. Abo, H. Tatsu, Simulation model based on non-
Y. Liu, Y.L. Zhao, D.L. Han, Y. Zhao, Y.G. Yang, W. Wu, Single-cell RNA-seq
Newtonian fluid mechanics applied to the evaluation of the embolic effect of
highlights intra-tumoral heterogeneity and malignant progression in pancreatic
emulsions of iodized oil and anticancer drug, Cardiovasc. Intervent. Radiol. 23
ductal adenocarcinoma, Cell Res. 29 (2019) 725–738, https://doi.org/10.1038/
(2000) 285–290, https://doi.org/10.1007/s002700010070.
s41422-019-0195-y.
[28] W. Zhang, Y. Zhang, Y. He, X. Xu, X. Zhao, Oil density and viscosity affect emulsion
[8] K.T. Butterworth, S.J. McMahon, F.J. Currell, K.M. Prise, Physical basis and
stability and destabilization mechanism, J. Food Eng. 366 (2024) 111864, https://
biological mechanisms of gold nanoparticle radiosensitization, Nanoscale 4 (2012)
doi.org/10.1016/j.jfoodeng.2023.111864.
4830–4838, https://doi.org/10.1039/C2NR31227A.
[29] C.H. Lee, V. Moturi, Y. Lee, Thixotropic property in pharmaceutical formulations,
[9] S. Bonvalot, P.L. Rutkowski, J. Thariat, S. Carrère, A. Ducassou, M.-P. Sunyach,
J. Control. Release 136 (2009) 88–98, https://doi.org/10.1016/j.
P. Agoston, A. Hong, A. Mervoyer, M. Rastrelli, V. Moreno, R.K. Li, B. Tiangco, A.
jconrel.2009.02.013.
C. Herraez, A. Gronchi, L. Mangel, T. Sy-Ortin, P. Hohenberger, T. de Baère, A. Le
[30] S. D’Souza, A review of in vitro drug release test methods for Nano-sized dosage
Cesne, S. Helfre, E. Saada-Bouzid, A. Borkowska, R. Anghel, A. Co, M. Gebhart,
forms, Adv. Pharm. 2014 (2014) 1–12, https://doi.org/10.1155/2014/304757.
G. Kantor, A. Montero, H.H. Loong, R. Vergés, L. Lapeire, S. Dema, G. Kacso,
[31] L. de Sousa Cavalcante, G. Monteiro, Gemcitabine: metabolism and molecular
L. Austen, L. Moureau-Zabotto, V. Servois, E. Wardelmann, P. Terrier, A.J. Lazar, J.
mechanisms of action, sensitivity and chemoresistance in pancreatic cancer, Eur. J.
V.M.G. Bovée, C. Le Péchoux, Z. Papai, NBTXR3, a first-in-class radioenhancer
Pharmacol. 741 (2014) 8–16, https://doi.org/10.1016/j.ejphar.2014.07.041.
hafnium oxide nanoparticle, plus radiotherapy versus radiotherapy alone in
[32] B. Ewald, D. Sampath, W. Plunkett, H2AX phosphorylation marks gemcitabine-
patients with locally advanced soft-tissue sarcoma (Act.In.Sarc): a multicentre,
induced stalled replication forks and their collapse upon S-phase checkpoint
phase 2–3, randomised, controlled trial, Lancet Oncol 20 (2019) 1148–1159,
abrogation, Mol. Cancer Ther. 6 (2007) 1239–1248, https://doi.org/10.1158/
https://doi.org/10.1016/S1470-2045(19)30326-2.
1535-7163.Mct-06-0633.
[10] A.F. Bagley, E.B. Ludmir, A. Maitra, B.D. Minsky, G. Li Smith, P. Das, A.C. Koong,
[33] S.I. Barroso, A. Aguilera, Detection of DNA double-Strand breaks by γ-H2AX
E.B. Holliday, C.M. Taniguchi, M.H.G. Katz, E.P. Tamm, R.A. Wolff, M.J. Overman,
Immunodetection, Methods Mol. Biol. 2153 (2021) 1–8, https://doi.org/10.1007/
S. Patel, M.P. Kim, C.D. Tzeng, N. Ikoma, M.S. Bhutani, E.J. Koay, NBTXR3, a first-
978-1-0716-0644-5_1.
in-class radioenhancer for pancreatic ductal adenocarcinoma: report of first patient
[34] V. Yazbeck, E. Alesi, J. Myers, M. H. Hackney, L. Cuttino, D. A. Gewirtz, Chapter
experience, Clin. Transl. Radiat. Oncol. 33 (2022) 66–69, https://doi.org/
one - an overview of chemotoxicity and radiation toxicity in cancer therapy. In
10.1016/j.ctro.2021.12.012.
Advances in Cancer Research, Gewirtz, D. A.; Fisher, P. B., Eds. Academic Press: U.
[11] J. Xie, L. Gong, S. Zhu, Y. Yong, Z. Gu, Y. Zhao, Emerging strategies of
S.A, 2022; pp. 1–27.
nanomaterial-mediated tumor Radiosensitization, Adv. Mater. 31 (2019)
[35] C. Verry, S. Dufort, B. Lemasson, S. Grand, J. Pietras, I. Troprès, Y. Crémillieux,
e1802244, https://doi.org/10.1002/adma.201802244.
F. Lux, S. Mériaux, B. Larrat, J. Balosso, G. Le Duc, E.L. Barbier, O. Tillement,
[12] C. Hoffmann, V. Calugaru, E. Borcoman, V. Moreno, E. Calvo, X. Liem, S. Salas,
Targeting brain metastases with ultrasmall theranostic nanoparticles, a first-in-
B. Doger, T. Jouffroy, X. Mirabel, J. Rodriguez, A. Chilles, K. Bernois, M. Dimitriu,
human trial from an MRI perspective, Sci. Adv. 6 (2020), https://doi.org/10.1126/
N. Fakhry, S.W. Hee Kam, C. Le Tourneau, Phase I dose-escalation study of
sciadv.aay5279 eaay5279-88.
NBTXR3 activated by intensity-modulated radiation therapy in elderly patients
[36] P.R. Rauta, Y. Mackeyev, K. Sanders, J.B.K. Kim, V.V. Gonzalez, Y. Zahra, M.
with locally advanced squamous cell carcinoma of the oral cavity or oropharynx,
A. Shohayeb, B. Abousaida, G.V. Vijay, O. Tezcan, P. Derry, A.V. Liopo, E.
Eur. J. Cancer 146 (2021) 135–144, https://doi.org/10.1016/j.ejca.2021.01.007.
R. Zubarev, R. Carter, P. Singh, S. Krishnan, Pancreatic tumor microenvironmental
[13] Z. Zhang, L. Zhou, N. Xie, E.C. Nice, T. Zhang, Y. Cui, C. Huang, Overcoming cancer
acidosis and hypoxia transform gold nanorods into cell-penetrant particles for
therapeutic bottleneck by drug repurposing, Signal Transduct. Target. Ther. 5
potent radiosensitization, Sci. Adv. 8 (2022) eabm9729, https://doi.org/10.1126/
(2020) 113, https://doi.org/10.1038/s41392-020-00213-8.
sciadv.abm9729.
[14] J.M. Idee, B. Guiu, Use of Lipiodol as a drug-delivery system for transcatheter
[37] D. Yu, Y. Zha, Z. Zhong, Y. Ruan, Z. Li, L. Sun, S. Hou, Improved detection of
arterial chemoembolization of hepatocellular carcinoma: a review, Crit. Rev.
reactive oxygen species by DCFH-DA: new insight into self-amplification of
Oncol. Hematol. 88 (2013) 530–549, https://doi.org/10.1016/j.
fluorescence signal by light irradiation, Sensor Actuat. B-Chem. 339 (2021)
critrevonc.2013.07.003.
129878, https://doi.org/10.1016/j.snb.2021.129878.
[15] S. Zhu, Y. Liao, C. Gu, D. Wang, H. Yan, L. Gao, D. Feng, Z. Gu, Clinically used
[38] M. Kanehisa, M. Furumichi, Y. Sato, M. Kawashima, M. Ishiguro-Watanabe, KEGG
lipiodol as an effective radioenhancer, Nano Today 56 (2024) 102279, https://doi.
for taxonomy-based analysis of pathways and genomes, Nucleic Acids Res. 51
org/10.1016/j.nantod.2024.102279.
(2023) D587–d592, https://doi.org/10.1093/nar/gkac963.
[16] T. Okusaka, M. Nakamura, M. Yoshida, M. Kitano, K. Uesaka, Y. Ito, J. Furuse, K.
[39] W. Kim, S. Lee, D. Seo, D. Kim, K. Kim, E. Kim, J. Kang, K.M. Seong, H. Youn,
Hanada, K. Okazaki, O. B. O. T. C. F. R. O. C. G. F. P. C. O. T. J. P. Society, Clinical
B. Youn, Cellular stress responses in radiotherapy, Cells 8 (2019), https://doi.org/
Practice Guidelines for Pancreatic Cancer 2019 From the Japan pancreas society: a
10.3390/cells8091105.
synopsis, Pancreas 49 (2020) 326–335. doi:https://doi.org/10.1097/m
[40] M.A. Morgan, L.A. Parsels, J. Maybaum, T.S. Lawrence, Improving gemcitabine-
pa.0000000000001513.
mediated radiosensitization using molecularly targeted therapy: a review, Clin.

252
S. Zhu et al. Journal of Controlled Release 374 (2024) 242–253

Cancer Res. 14 (2008) 6744–6750, https://doi.org/10.1158/1078-0432.Ccr-08- [44] J.W. Chung, J.H. Park, J.G. Im, J.K. Han, M.C. Han, Pulmonary oil embolism after
1032. transcatheter oily chemoembolization of hepatocellular carcinoma, Radiology 187
[41] U.S. Srinivas, B.W.Q. Tan, B.A. Vellayappan, A.D. Jeyasekharan, ROS and the DNA (1993) 689–693, https://doi.org/10.1148/radiology.187.3.8388567.
damage response in cancer, Redox Biol. 25 (2019) 101084, https://doi.org/ [45] H. Yarmohammadi, A.G. Aguirre, L. Schook, E. Ziv, J. Erinjeri, K. Brown,
10.1016/j.redox.2018.101084. G. Srimathveeravalli, E. O’Reilly, P. Allen, J. Haaga, S. Solomon, F. Boas,
[42] I. Aier, P.K. Varadwaj, Understanding the mechanism of cell death in gemcitabine Treatment of pancreatic cancer by intra-arterial injection of an emulsion of
resistant pancreatic ductal adenocarcinoma: a systems biology approach, Curr. Lipiodol and bumetanide (an anti-glycolytic drug) in a porcine model: initial
Genomics 20 (2019) 483–490, https://doi.org/10.2174/ results, J. Vasc. Interv. Radiol. 28 (2017) S8–S9, https://doi.org/10.1016/j.
1389202920666191025102726. jvir.2016.12.598.
[43] M. Zhu, X. Li, S. Wang, W. Guo, X. Li, Characterization of Radiotherapy Sensitivity
Genes by Comparative Gene Set Enrichment Analysis, Cham, Springer
International Publishing: Cham, 2018, pp. 205–216.

253