EVIDENCE-BASED REVIEW

Mechanism of Action, Pharmacokinetics,

Efficacy, and Safety of Transarterial
Therapies Using Ethiodized Oil: Preclinical
Review in Liver Cancer Models
Ron C. Gaba, MD, Regina M. Schwind, BS, and Sebastien Ballet, PhD

ABSTRACT

Purpose: To systematically review mechanism of action, pharmacokinetics (PKs), efficacy, and safety of ethiodized oil–based
locoregional therapy (LRT) for liver cancer in preclinical models.
Materials and Methods: A MEDLINE search was performed from 1988 to 2016. Search terms included hepatocellular carcinoma
(HCC), HCC, liver-cell carcinoma, liver, hepatic, hepatocarcinoma, transarterial or chemoembolization, TACE, animal, Lipiodol,
Ethiodol, iodized oil, and/or poppy-seed oil. Inclusion criteria were: publication in a peer-reviewed journal, an accepted animal model,
and PK/safety/efficacy data reported. Exclusion criteria were: inadequate PK, safety, or efficacy data; anticancer drug name/dose not
available; and article not in English. Outcomes included intratumoral anticancer drug uptake, PKs, tolerance, tumor response, and
survival.
Results: Of 102 identified articles, 49 (49%) met the inclusion criteria. Seventeen, 35, and 2 articles used rat, rabbit, and pig models.
Mechanism of action was investigated in 11 articles. Eleven articles reported drug uptake, PK, and tolerance data, showing 0.5%–9.5%
of injected chemotherapy dose in tumor. Tumor-to-liver drug distribution ratios were 2–157. Toxicology data across 6 articles showed
transient liver laboratory level elevations 1 day after LRT. There was no noteworthy liver or extrahepatic histologic damage. Nine
articles reported tumor response, with 0%–30% viable tumor and –10% to –38% tumor growth at 7 days after LRT. Two articles reported
survival, showing significantly longer survival after LRT vs untreated controls (56/60 d vs 33/28 d). Several articles described ethiodized
oil mixed with radiopharmaceutical (n ¼ 7), antiangiogenic (n ¼ 6), gene (n ¼ 6), nanoembolic (n ¼ 5), immune (n ¼ 2), or other novel
(n ¼ 1) agents.
Conclusions: Animal studies show preferential tumor uptake of anticancer agent, good hepatic/systemic tolerance, high tumor
response, and enhanced survival after ethiodized oil–based LRT.

ABBREVIATIONS

DEE ¼ drug-eluting embolic, FdUrd-C8 ¼ 30 ,50 -dioctanoyl-5-fluoro-20 -deoxyuridine, HCC ¼ hepatocellular carcinoma, IL ¼ inter-
leukin, LRT ¼ locoregional therapy, MN-16ET ¼ N-[2-(triphenylmethyl)thioethyl]-3-aza-19-ethyloxycarbonyl-3-[2-(triphenylmethyl)
thioethyl]octadecanoate, ODN ¼ oligodeoxynucleotide, PK ¼ pharmacokinetic, pMNC ¼ porous magnetic nanocluster, siRNA ¼
small interfering RNA, TNF ¼ tumor necrosis factor, VEGF ¼ vascular endothelial growth factor

Conventional transarterial chemoembolization refers to

infusion of chemotherapeutic agents as an emulsion with
From the Department of Radiology, Division of Interventional Radiology ethiodized oil (Lipiodol Ultra-Fluid; Guerbet, Villepinte,
(R.C.G., R.M.S.), University of Illinois Hospital and Health Sciences System, France) followed by embolization with particulate agents
1740 W. Taylor St., MC 931, Chicago, IL 60612; and Guerbet Research and (1). During conventional chemoembolization, ethiodized oil
Innovation Department (S.B.), Aulnay-sous-Bois, France. Received June 16,
2017; final revision received September 20, 2017; accepted September 27, serves to emulsify and carry chemotherapeutic agents to
2017. Address correspondence to R.C.G.; E-mail: rgaba@uic.edu tumor (2), facilitate intracellular drug entry (3), and embo-
lize tumor microcirculation (4). Since the first reports of
R.C.G. receives grants from Guerbet (Villepinte, France). Neither of the other
authors has identified a conflict of interest. ethiodized oil chemoembolization in the early 1990s, the
clinical benefit of conventional chemoembolization has been
© SIR, 2017
widely validated in medical practice (5). Nonetheless, the
J Vasc Interv Radiol 2017; ▪:1–12 broadening focus of cancer therapeutic agents on minimally
https://doi.org/10.1016/j.jvir.2017.09.025 invasive, locoregional, and targeted treatments, as well as
2 ▪ Transarterial Therapies with Ethiodized Oil in Liver Cancer Models Gaba et al ▪ JVIR

clinical advances in conventional chemoembolization, have

spurred preclinical research to explore the mechanism of
action of ethiodized oil as a drug delivery system and to
discover new anticancer agent and ethiodized oil formula-
tions. Such preclinical studies rely heavily on predictive
animal cancer models to allow clinical translation for liver
cancer treatment.
A comprehensive understanding of the available mecha-
nistic, pharmacokinetic (PK), efficacy, and safety data of the
reported ethiodized oil–based transarterial therapies in the
treatment of liver cancer in animal models is therefore vital
for the development of new human trials. This study was Figure 1. Flow chart delineating manuscript identification.
therefore undertaken to systematically review the available
literature on the mechanism of action, PKs, efficacy, and
safety of ethiodized oil-based locoregional therapies (LRTs) Outcome Measures
of liver cancer in preclinical models. Data were compiled, and relevant outcomes—including
intratumoral anticancer drug uptake, PK data, tolerance,
tumor response, and survival—were tabulated on a per-
MATERIALS AND METHODS study basis. Studies reporting outcomes of conventional
Data Sources and Selection Process chemoembolization with standard therapeutic agents (eg,
A literature search was performed by a single reviewer doxorubicin) were reported distinctly from those that
(S.B.) using the MEDLINE database spanning the period employed innovative agents.
from 1988 to 2016. The abstracts of identified publications
were screened by two reviewers (S.B. and R.C.G.) for RESULTS
possible inclusion based on specific eligibility criteria,
which were cooperatively developed by all authors. Eligible Literature Search Results
articles were reviewed in full by all authors and assessed for Literature search yielded a total of 102 articles. Of these, 49
inclusion. Interreviewer discrepancies were resolved (48%) met inclusion criteria (2,4,6–52), of which 17 (35%),
through consensus discussion. References within studies 35 (71%), and 2 (4%) articles used rat, rabbit, and pig models
that met selection criteria were searched for other potentially (some articles used more than one animal species). A flow
relevant studies. chart of manuscript identification is presented in Figure 1.
The included studies investigated mechanism of action
(n ¼ 11; 23%), ethiodized oil/chemotherapy PKs and
Search Terms efficacy (n ¼ 16; 33%), ethiodized oil/chemotherapy safety
The following search terms—jointly conceived by all (n ¼ 6; 13%), therapeutic radionuclide administration
authors—were used to identify potential articles: “hepato- (n ¼ 7; 15%), antiangiogenic agent delivery (n ¼ 6; 13%),
cellular carcinoma” or “liver-cell carcinoma” and “trans- gene therapy (n ¼ 6; 13%), nanoembolization (n ¼ 5;
arterial chemoembolization” or “chemoembolization” or 10%), immunoembolization (n ¼ 2; 4%), and novel
“intra-arterial injection” or “transarterial injection” or chemotherapy agent LRT (n ¼ 1; 2%).
“hepatic artery injection” and “cell line model” or “mice
model” or “rat model” or “rabbit model” or “pig model.” All
search terms were also combined with “Lipiodol,” “Lipiodol Mechanism of Action
Ultra-Fluid,” “Ethiodol,” “iodized oil,” and “poppy-seed oil.” Ethiodized oil–mediated drug delivery.—The intravascular
distribution of ethiodized oil after intra-arterial infusion
was described in a rabbit model by de Baere et al in 1995
Inclusion and Exclusion Criteria (6): oil droplets showed a propensity to pass through the
English language studies published in peer-reviewed jour- largest arteries (diameter 30 larger than oil droplets) at a
nals that reported PK parameters (plasma anticancer drug bifurcation, without passage into smaller vessels. Within
level, drug tumor uptake, drug tumor-to-liver ratio), efficacy 12–18 seconds after infusion, fourth- or fifth-order
(tumor response, survival), or safety (serum hepatic marker bifurcations were reached by intact, elongated oil drop-
level, liver histologic examination) data of ethiodized oil– lets to penetrate terminal arterioles and microcapillaries
based LRT in accepted animal models with liver cancer were (6,7). This selectivity accounted for the tendency of
included. Exclusion criteria included ethiodized oil sus- ethiodized oil droplets to preferentially “select” larger tu-
pension (given the lack of correspondence to clinical con- mor neovasculature.
ventional chemoembolization), inadequate safety or efficacy When ethiodized oil was injected into the hepatic artery in
data, anticancer drug name and/or dose not available, mouse rat, rabbit, and swine, oil phase also dose-dependently
model (not widely accepted), and article not published in appeared in the terminal portal venules (2,4,7–9) (Fig 2).
English. Histologic and in vivo microscopy examinations of livers
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017 3

Figure 2. Serial in vivo micrographs (magnification 500) of rat liver 4 minutes after injection of 0.2 mL/kg ethiodized oil into hepatic
artery demonstrating oil droplet (O) accumulation in terminal portal venule (PV). Reprinted from Kan et al (4) with permission from
Elsevier.

showed that ethiodized oil accumulated in portal venules,

sinusoids, and peribiliary plexus after arterial injection of
oil-in-water and water-in-oil emulsions (2,4,7,9,10). If oil
droplets were larger than sinusoids (5–8 μm), they blocked
the sinusoids, stopping or reducing intratumoral blood flow
(2,9) (Fig 3). Oil droplets smaller than sinusoids (1–3 μm)
circulated into the border tumor vasculature.
Regarding intratumoral retention of ethiodized oil/
chemotherapy emulsion, de Baere et al (11) showed that
water-in-oil emulsions were more densely retained within
tumors than oil-in-water emulsions. When ethiodized oil
and doxorubicin/iodinated contrast material were mixed at a
4:1 ratio, in vitro analysis has shown approximately 30% Figure 3. In vivo liver micrograph of Fischer rat with mammary
chemotherapy release after 8 hours (12). cancer metastasis. Rat received hepatic arterial injection of
ethiodized oil emulsion (arrowheads), which lodged within
Based on its rheologic properties, ethiodized oil also
tortuous intratumoral blood vessel. w ¼ water bubble. Reprinted
acts a transient embolic agent (6,7). The time for liver from Kan et al (2) with permission from Elsevier.
clearance of ethiodized oil is dose-dependent (15 d at 0.1
mL/kg to 60 d at 0.4 mL/kg in rat), and is longer than the
resonance (MR) imaging to successfully quantify ethiodized
time for the recovery of the hepatic microcirculation (9).
oil delivered via a transarterial approach to rat livers.
In vivo microscopy demonstrated that Kupffer cells
endocytose ethiodized oil droplets (9). The droplets are
cleared from portal vessels in 2–3 days and from sinusoids PK and Efficacy Overview
in 7 days, with total disappearance at 15 days in a healthy Quantitative PK and efficacy results of ethiodized oil–based
rat model (7). LRT with the use of standard chemotherapy in animal
Quantification of LRT.—In 2012, Gaba et al (13) studied models are summarized in Table 1 (11–13,15–26,52).
the correlation between rabbit VX2 intratumoral ethiodized PK and Efficacy in Specific Investigations.—The thera-
oil and doxorubicin content. The investigators found no pre- peutic effects of anticancer drug dissolved in ethiodized oil
dictable correlation, but the study was limited by the use of an was first studied in a rabbit VX2 model by Fukushima et al
oil-in-water ethiodized oil/chemotherapy emulsion for che- in 1987 (15). In rabbits receiving increasing doses of 30 ,50 -
moembolization. In 2012, Yin et al (14) employed magnetic dioctanoyl-5-fluoro-20 -deoxyuridine (FdUrd-C8), tumor
 4 ▪ Transarterial Therapies with Ethiodized Oil in Liver Cancer Models
Table 1. Tumor Uptake and Response Data from Translational Animal Studies of Intraarterial Ethiodized Oil–Based Formulations (11–13,15–26,52)

Study, Year Model Catheter Site Treatments Tumor Uptake Tumor Response
Fukushima et al Rabbit VX2, PHA FdUrd-C8 (30–100 mg)/EO solution – Day 7: tumor growth, 34.8, 14.9, –2.4, and
(15), 1987 N ¼ 8-13 (0.2 mL) –10.4% in groups treated with 30, 50,
70, and 100 mg FdUrd-C8; tumor
growth of untreated animals, 636%.
Kawaguchi et al Rabbit VX2, PHA FdUrd-C8 (100 mg)/EO solution Drug tumor-to-liver ratios of 17, 30, and –
(16), 1988 N¼4 157 at 2, 5, and 24 h; 250 μg FdUrd-C8/
g tumor 72 h after treatment and not
detectable in liver
Konno (17), 1990 Rabbit VX2, PHA PK, 14C-labeled DOX (DOX/EO vs DOX Intratumoral DOX: DOX/EO, 20,358– Tumor size at treatment/1 wk after
N ¼ 96 aqueous solution in EO); treatment: 28,066 dpm/g immediately; 2,505– treatment: group 1, 217/1,417 mm2;
group 1, sham; group 2, 0.2 mL EO; 12,800 dpm/g at 7 d; DOX aqueous group 2, 263 /922 mm2; group 3,
group 3, 0.2 mL mitomycin C/EO (4 solution in EO, 9,606–15,309 dpm/g 261/175 mm2 (complete necrosis in
mg/mL); group 4, 0.2 mL mitomycin immediately; 0–77 dpm/g at 7 d 8/9); group 4, 209/375 mm2 (complete
C aqueous solution (4 mg/mL) in 0.2 necrosis in 2/16); group 5, 254/159
mL EO; group 5, 0.2 mL aclarubicin C/ mm2 (complete necrosis in 6/8); group
EO (12.5 mg/mL); group 6, 0.2 mL 6, 239/405 mm2 (complete necrosis in
aclarubicin aqueous solution (12.5 1/12); group 7, 243/204 mm2 (complete
mg/mL) in 0.2 mL EO; group 7, 0.2 mL necrosis in 5/7); group 8, 268/486 mm2
DOX/EO (2 mg/mL); group 8, 0.2 mL (complete necrosis in 0/7)
DOX aqueous solution (2 mg/mL) in
0.2 mL EO
125
de Baere et al (11), Rabbit VX2, Hepatic artery I-labeled EO emulsions (0.1 mL): Tumor/liver ratio of 3.6–10.2 at 4 d –
1996 N¼4 water-in-oil, oil-in-water, small
droplets (10–40 μm), large droplets
(30–120 μm)
Yoon et al (18), Rabbit VX2, LHA Paclitaxel (low dose 1 mg or high dose Paclitaxel peak in tumor at 1 and 3 d Tumor growth at 7 d, 3.3% for high-dose
2003 N¼6 4 mg)/EO (0.4 mL) solution (> 80 μg/g tumor) and gradual release paclitaxel, 18.7% for low-dose
in liver at 7 d (< 10 μg/g tumor) paclitaxel, and 68% in untreated
animals; residual viable tumor at 7 d,
2.8% for high dose vs 12.7% for low
dose vs 31.1% for untreated animals
Yoon et al (19), Rabbit VX2, LHA DOX (2 mg)/EO (0.4 mL) emulsion or – Mean tumor growth at 1, 2, and 3 wk,
2004 N ¼ 15-21 EO alone (0.4 mL) –3%, –5%, and 29% with DOX/EO,
lower than EO group (46%, 145%,
283%); mean residual viable tumor, 3%
in DOX/EO group vs 30% in EO group
Baumgarten, Gaba, Rabbit VX2, ND DOX (2 mg)/EO (1 mL) emulsions At 30 min in peripheral tumor tissue, –
and van Breemen N¼4 DOX 7.5 μg/mL, iodine 498 mg/mL; in
(20), 2012 core tumor tissue, DOX 6.9 μg/mL,

Gaba et al ▪ JVIR
iodine 164 mg/mL; DOX 5% ID and EO
1% ID at 30 min (assuming that tumor
of 1.9 cm diameter corresponds to
volume of 3.6 mL for each tumor)
continued
 Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017
Table 1. Tumor Uptake and Response Data from Translational Animal Studies of Intraarterial Ethiodized Oil–Based Formulations (11–13,15–26,52) (continued)

Study, Year Model Catheter Site Treatments Tumor Uptake Tumor Response
Becker et al (21), Rat N1-S1, GDA 99m
Tc-EO At 24 h, 9.5% ID/g of tumor, 9.3% ID/g of –
2012 N ¼ 6-9 liver, 2.6% ID/g of lung, tumor-to-liver
ratio of 1.13; at 72 h, 7.5% ID/g of
tumor, 4.2% ID/g of liver, 2.4% ID/g of
lung, tumor-to-liver ratio of 2.16
Gaba et al (13), Rabbit VX2, PHA or LHA DOX (mean 1.8 mg)/EO (mean 0.9 mL ¼ At 30 min, 6% ID of DOX (mean 15.8 μg/ –
2012 N¼6 432 mg iodine) emulsion (ratio 2:1) mL) and 0.55% ID of iodine in tumor
(mean 335 μg/mL;
assuming that tumor of 1.9 cm
diameter corresponds to volume of 3.6
mL for each tumor)
Ueda et al (22), Rabbit VX2, MHA Epirubicin (1.5 mg)/EO emulsion – Mean tumor growth at 7 d, þ21.4%;
2012 N¼5 (0.3 mL), ratio 1:1 viable proportion of tumors at 7 d,
32%–59%
Liang et al (23), Rabbit VX2, PHA Group 1, IV DOX alone; group 2, IA DOX penetration in tumor at 4 h, group 1, –
2013 N¼8 DOX alone; group 3, IA DOX followed 7.36 μm; group 2, 32.68 μm; group 3,
by 150–250 μm PVA; group 4, IA 72.37 μm; group 4, 81.16 μm; DOX
DOX/EO emulsion followed by PVA fluorescence at 4 h, group 1, 75 counts;
150–250 μm embolization; injected group 2, 1,137 counts; group 3, 3,190
doses, 30 mg DOX per animal, 0.4 mL counts; group 4, 3,681 counts
EO, 0.2–0.4 mL PVA
Lin et al (24), 2013 Rat N1-S1, GDA Group 1, 188Re-MN-16ET- EO – Tumor size at 8 wk, group 1, 100% good
N¼4 (0.2 mCi/7.4 MBq); group 2, responders with mean decrease –56%;
epirubicin (0.5 mg/kg)/EO (0.1 mL) group 2, 50% good responders with
emulsion; group 3, saline solution mean decrease –18%; group 3, no
responders; survival time, > 56 d for
groups 1 and 2, 32.7 d for group 3
Tomozawa et al Rabbit VX2, LHA Miriplatin (6 mg)/EO (0.3 mL) emulsion Platinum content at 24 h, 4.2–16.4 μg/g Decrease of tumor growth at 1 wk, from
(25), 2013 N¼5 plus contrast medium (MLC) or saline tumor, 1.2–4.2 μg/g liver, tumor-to- 53% to 67% after MLS or MLC
solution (MLS); 4 ratios for each liver ratio of 1.9–9.3 treatment vs saline; no difference
emulsion type between MLS and MLC
Lilienberg et al (26), Pig, N ¼ 14 Hepatic artery Group 1, EO/DOX, DOX 3.1 mg injected, AUC0–6 (min$nM), femoral vein, 180 (oil/ –
2014 water:oil volume ratio 1:3.33; group DOX 3.1 mg), 88 (DEE-DOX), 230 (oil/
2, DEE-DOX, DOX 50 mg injected, DOX 22 mg); portal vein, 340 (oil/DOX
100–300-μm DEE; group 3, EO/DOX, 3.1 mg), 130 (DEE-DOX), 100 (oil/DOX
DOX 22 mg injected, water:oil 22 mg); hepatic vein, 230 (oil/DOX 3.1
volume ratio 1:3.33 mg), 260 (DEE-DOX), 400 (oil/DOX 22
mg); bioavailability, hepatic vein, 85%
(oil/DOX 3.1 mg), 24% (DEE-DOX), 94%
(oil/DOX 22 mg); femoral vein, 29%
(oil/DOX 3.1 mg), 14% (DEE-DOX), 36%
(oil/DOX 22 mg)
continued

5
6 ▪ Transarterial Therapies with Ethiodized Oil in Liver Cancer Models Gaba et al ▪ JVIR

volume at 10 d, 1,774 mm3 for DOX/EO

Tumor growth ratio at 10 d, 4.7 for DOX/
and no viable tumor in 3 rabbits in DEE

EO group, 3.4 for lidamycin/EO; tumor

(triphenylmethyl)thioethyl]octadecanoate (MN-16ET)-ethiodized oil; EO ¼ ethiodized oil; FdUrd-C8 ¼ 30 ,50 -dioctanoyl-5-fluoro-20 -deoxyuridine; GDA ¼ gastroduodenal artery; IA ¼
DEE ¼ drug-eluting embolic; DOX ¼ doxorubicin; dpm ¼ disintegrations per minute; 88Re-MN-16ET ¼ 188Re-labeled N-[2-(triphenylmethyl)thioethyl]-3-aza-19-ethyloxycarbonyl-3-[2-

intraarterial; ID ¼ injected dose; IV ¼ intravenous; LHA ¼ left hepatic artery; MHA ¼ middle hepatic artery; ND ¼ not determined; PHA ¼ proper hepatic artery; PVA ¼ polyvinyl alcohol;
growth was drastically suppressed at 7 days after treatment.
At 7 d, no viable tumor in groups A and

group vs 1,145 mm3 for lidamycin/EO

B; viable tumor observed in 2 rabbits
Ethiodized oil was detected at the tumor site in all treated
animals by using ex vivo liver radiography. Histologic ex-
amination revealed a dose-dependent larger necrotic portion,
Tumor Response

and no viable tumor was detected in animals treated with the

highest doses of FdUrd-C8.
The same research group (16) published data after in-
Table 1. Tumor Uptake and Response Data from Translational Animal Studies of Intraarterial Ethiodized Oil–Based Formulations (11–13,15–26,52) (continued)

jection of a fixed FdUrd-C8 concentration per rabbit and

noted slower elimination of FdUrd-C8 from tumor (half-life,
15.8 h) than from nontumorous regions (half-life, 3.8-4.2 h),
resulting in selective intratumoral FdUrd-C8 accumulation.
group

group
The tumor-to-liver ratios reached 17, 30, and 157 at 2, 5,
and 24 hours after treatment. The tumor-to-blood ratio
ranged from 870 to 5,400 over a period of 15–1,440 minutes
in DEE group; DOX in liver, 149.9 ng/g
tissue in group A, 25.1 ng/g in group B,
group A, 1.7 μg/g in group B, 10.2 μg/g
At 7 d, DOX in tumor, 6.7 μg/g tissue in

after infusion.
tumor-to-liver ratio, 45.3 in group A,
and 198.1 ng/g in DEE group; DOX

65.2 in group B, 51.7 in DEE group

Subsequent studies reinforced the anticancer efficacy and

drug delivery properties of ethiodized oil (13,17,20,23).
Konno (17) investigated intratumoral PKs and tumor
Tumor Uptake

response across various ethiodized oil/chemotherapy for-

mulations, demonstrating high intratumoral accumulation
–

associated with VX2 tumor necrosis (Fig 4). Intra-arterial

injection of doxorubicin/ethiodized oil emulsion in a rab-
bit VX2 model resulted in intratumoral drug concentration
corresponding to 6% of injected dose (13). Another VX2
study showed that doxorubicin tumor levels were equally
measured in peripheral and core tumor, corresponding to 5%
of the injected dose (20). Liang et al (23) compared the
effect of different LRTs in rabbit VX2 tumors on the dis-
animal of 2.4 mg in all treated groups
DOX (6 mg)/EO and lidamycin (0.3 mg)/
with 1.25 mL EO; group C, DEE 100–
300 μm or 300–500 μm loaded with
iodinated contrast agent emulsified

DOX at 37.5 mg/mL; DOX dose per

ratio 1:1, 10 mg DOX in 1.25 mL of
Group A, ratio 1:4, 10 mg DOX in 0.5

emulsified with 2 mL EO; group B,

tribution of doxorubicin in four treatment groups: intrave-

nous doxorubicin (group 1), intra-arterial doxorubicin
mL of iodinated contrast agent

(group 2), intra-arterial doxorubicin followed by particle

embolization (group 3), and intra-arterial doxorubicin/
Treatments

ethiodized oil followed by particle embolization (group 4).

Intra-arterial treatment led to a significant increase in
intratumoral doxorubicin penetration compared with group
EO emulsions

1. Four hours after treatment, groups 3 and 4 had significant

increases in drug penetration compared with group 2. A
trend of greater doxorubicin penetration was observed in
group 4.
In 1996, de Baere et al (11) investigated the tumor uptake
of different ethiodized oil–based emulsions, demonstrating
Catheter Site

preferential tumor uptake of radiolabeled ethiodized oil

compared with nontumorous liver (ratios of 3.8–10.3) in a
VX2 model. Greater tumor uptake was observed with water-
LHA

LHA

in-oil emulsions composed of 30–120-μm droplets. Becker

et al (21) investigated the influence of ethiodized oil on
Rabbit VX2,

Rabbit VX2,

tumor retention in a rat N1-S1 model, revealing that intra-

Model

N ¼ 10

TNF ¼ tumor necrosis factor.

tumoral uptake was maintained for at least 3 days after intra-

N¼5

arterial administration whereas nontumorous liver uptake

was halved from 24 to 72 hours after administration; tumor-
to-liver ratio increased between the first day (ratio of 1.13)
Choi et al (12), 2014

and the third day (ratio of 2.16) after treatment.

Zhong et al (52)

In a preclinical study in 12 healthy pigs, PK data of

Study, Year

doxorubicin in three blood compartments (portal, hepatic,

and femoral vein) were measured after arterial injection of
2015

ethiodized oil/doxorubicin mixture or doxorubicin-loaded

drug-eluting embolic (DEE) particles (26). Greater
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017 7

Innovative Anticancer Agents Delivered

by Ethiodized Oil
Delivery of therapeutic radionuclide.—Rhenium-188
(188Re) is a therapeutic radionuclide with a similar energy
(2.2 MeV) characteristic of clinically employed yttrium-90,
but with a shorter physical half-life (17.4 h vs 64.3 h).
Table 3 summarizes studies of 188Re-ethiodized oil LRT
(19,24,28–32).
In a 1996 study by Wang et al (28), injection of
188
Re-ethiodized oil in a rat N1-S1 model showed high
Figure 4. Low-kilovoltage radiograph (magnification 2) of
intratumoral radioactivity, with a biological half-life of
resected VX2 specimen 7 days after arterial injection treated with 122.9 hours. Radioactivity in nontumorous liver was
mitomycin C and ethiodized oil shows persistent ethiodized oil lower, with a shorter biological half-life of 31.7 hours. The
accumulation; pathologic assessment (not shown) revealed tumor-to-liver ratio demonstrated preferential tumor reten-
complete tumor necrosis. Reprinted from Konno et al (17) with tion of radiolabeled ethiodized oil with a higher radiation
permission from John Wiley & Sons.
dose (28).
Garin et al (29) reported the use of an ethiodized oil Super-
systemic exposure (in the femoral vein) of doxorubicin was Six Sulfur 188Re-ligand complex, and injection of 5.55 MBq
observed following ethiodized oil delivery compared with of this complex in a rat N1-S1 model revealed good tumor
DEE delivery, but with a higher posthepatic (ie, hepatic retention with a high tumor-to-liver ratio (mean, 3.8). Simi-
vein) bioavailability, indicating that release from ethiodized larly, Luo et al (30) measured a high initial 188Re tumor uptake
oil was more rapid and extensive than from DEE particles after the injection of 188Re-ethyl cysteinate dimer/ethiodized
even if the injected doxorubicin dose was approximately 17- oil in a rat N1-S1 model, with a high tumor-to-liver ratio (30).
fold lower with ethiodized oil. Ethiodized oil appeared to Yoon et al (19) developed a 188Re-4-hexadecyl-2,2,9,9-
rapidly distribute doxorubicin to the extravascular space of tetramethyl-4,7-diaza-1,10-decanethiol/ethiodized oil com-
the liver. plex, and injection in VX2 tumors yielded an intratumoral
More recently, a rabbit VX2 model was used to compare half-life of 16.2 hours with a tumor radiation dose of 147.7
in vivo PKs of doxorubicin and therapeutic response from Gy. Mean tumor growth at 1, 2, and 3 weeks and percentage of
different ethiodized oil emulsions and DEE agents (12). viable tumor afterward (3.4%, 7.6%, 11.1%, and 0.3%)
Animals were treated with a 4:1 volume ratio (oil:aqueous) were similar to those after intra-arterial doxorubicin/ethio-
ethiodized oil/doxorubicin emulsion (group A), 1:1 ratio dized oil treatment (3.2%, 5.3%, 29.0%, and 2.6%), and
ethiodized oil/doxorubicin emulsion (group B), or were lower than after treatment with intra-arterial ethiodized
doxorubicin-loaded DEE particles (group C). Whereas the oil alone (45.5%, 145.4%, 283.0%, and 30.1%).
maximum plasma concentrations of doxorubicin were In 2011, 188Re-labeled N-[2-(triphenylmethyl)thioethyl]-
different between animal groups (group A, 384.99 ng/mL; 3-aza-19-ethyloxycarbonyl-3-[2-(triphenylmethyl)thioethyl]
group B, 830.30 ng/mL; group C, 68.48 ng/mL), the plasma octadecanoate (MN-16ET)/ethiodized oil was evaluated
area-under-the-curve value—indicating total systemic (24,31,32). Administration of 188Re-MN-16ET without
exposure to doxorubicin—in group A (2,409 ng$min/mL) ethiodized oil in a rat N1-S1 model showed a decrease in
was significantly lower than in group B (4,612 ng$min/mL) radioactivity after treatment, along with sustained intra-
and was similar to that in group C (1,346 ng$min/mL). In all tumoral retention (31). In 2013, Huang et al (32) showed
groups, intratumoral doxorubicin concentrations were higher that intra-arterial injection of 188Re-MN-16ET/ethiodized
than in normal liver, with similar tumor-to-liver ratios (group oil in a rat N1-S1 model significantly decreased tumor
A, 45.3; group B, 65.2; group C, 51.7). Histologic exami- size, whereas tumors grew in untreated rats or those
nation revealed no viable tumor in all rabbits in the ethio- receiving ethiodized oil alone. All rats treated with 188Re-
dized oil chemoembolization groups, whereas viable tumor MN-16ET/ethiodized oil were still alive at 60 days,
was present in 2 rabbits in the DEE agent–treated group. whereas untreated and ethiodized oil–treated rats survived
28.2 and 47.7 days, respectively. Lin et al (24) confirmed
Safety Results that rat N1-S1 tumors receiving 188Re-MN-16ET/ethio-
Safety data associated with ethiodized oil–based LRTs are dized oil showed favorable tumor response, with mean
summarized in Table 2 (12,15,17–19,22,27). In rabbit size reduction of 56%. All rats receiving 188Re-MN-16ET/
studies, increases in serum aminotransferase levels was ethiodized oil or epirubicin/ethiodized oil survived over 8
observed at 1 day after treatment, with a return to baseline weeks, whereas survival of untreated rats was 32.7 days.
levels by 7 days after treatment. Histologic examination Delivery of Antiangiogenic Therapies.—In 2013, Gaba
revealed predominantly focal mild-to-moderate liver et al (33) tested the feasibility of intra-arterial sorafenib
infarction in rabbits treated with conventional or DEE che- LRT in rabbits, reporting successful delivery in 5 of 5
moembolization (12,17). animals (100%), with mean tissue sorafenib concentrations
 8 ▪ Transarterial Therapies with Ethiodized Oil in Liver Cancer Models
Table 2. Serum Hepatic Markers and Liver Histologic Data from Animal Treated with Hepatic Intraarterial Ethiodized Oil–Based Formulations (12,15,17–19,22,27)

Study, Year Model Catheter Treatments Serum Hepatic Markers Liver Histology
Site
Fukushima et al (15), 1987 Rabbit VX2, PHA FdUrd-C8 (30–100 mg)/EO solution Transient AST (100–500 U/L) and Necrosis and degeneration observed
N ¼ 8–13 (0.2 mL) ALT increase (60-400 U/L) at 1 d; in liver
no difference between treatments;
return to baseline at 5 or 7 d
Konno (17), 1990 Rabbit VX2, PHA Group 1, sham; group 2, 0.2 mL EO; – Group 3, necrosis of small area of
N ¼ 96 group 3, 0.2 mL mitomycin C/EO liver in 1/9; group 4, necrosis of
(4 mg/mL); group 4, 0.2 mL small area in 5/16, extensive
mitomycin C aqueous solution necrosis in 5/16; group 5, necrosis
(4 mg/mL) in 0.2 mL EO; group 5, of small area in 1/9, extensive
0.2 mL aclarubicin C/EO (12.5 mg/ necrosis in 3/9; group 6, necrosis
mL); group 6, 0.2 mL aclarubicin of small area in 3/15, extensive
aqueous solution (12.5 mg/mL) in necrosis in 3/15; group 7, no
0.2 mL EO; group 7, 0.2 mL DOX/ necrosis; group 8, necrosis of
EO (2 mg/mL); group 8, 0.2 mL small area in 3/7
DOX aqueous solution (2 mg/mL)
in 0.2 mL EO
Akashi et al (27), 1993 Rabbit VX2, PHA Group 1, DOX (2.8–3.8 mg) in Transient AST (300–600 UI/L) and –
N¼5 Urografin solution; group 2, EO ALT increase (< 400 UI/L) at 1 d; no
alone (0.3–1.2 mL); group 3, DOX difference between treatments;
(1 mg/kg) in 60% Urografin and EO return to baseline at 7 d
(0.6–2.4 mL)
Yoon et al (18), 2003 Rabbit VX2, LHA Paclitaxel (low dose 1 mg or high Transient ALT (650–1,100 U/L) and –
N¼6 dose 4 mg)/EO (0.4 mL) solution AST increase (400–900 U/L) at 1 d;
return to baseline at 5 or 7 d
Yoon et al (19), 2004 Rabbit VX2, LHA DOX (2 mg)/EO (0.4 mL) emulsion or Transient AST (500–600 U/L) and –
N ¼ 15–21 EO alone (0.4 mL) ALT increase (300–400 U/L) at 1 d;
no difference between treatments;
return to baseline at 3–7 d
Ueda et al (22), 2012 Rabbit VX2, N ¼ 5 MHA Epirubicin (1.5 mg)/EO emulsion (0.3 Transient AST (325 U/L) and ALT –
mL), ratio 1:1 increase (200 U/L) at 24 h; return to
baseline at 7 d
Choi et al (12), 2014 Rabbit VX2, N ¼ 5 LHA Group A, ratio 1:4, 10 mg DOX in 0.5 Transient AST (250–300 U/L) and Focal liver infarction in 60% of group
mL of iodinated contrast agent ALT increase (300–400 U/L) at 1 d; A, 100% of group B, and 20% of
emulsified with 2 mL EO; group B, no difference between treatments; group C
ratio 1:1, 10 mg DOX in 1.25 mL of return to baseline at 7 d
iodinated contrast agent
emulsified with 1.25 mL EO; group
C, DEE 100–300 μm or 300–500 μm

Gaba et al ▪ JVIR
loaded with DOX at 37.5 mg/mL;
DOX dose per animal, 2.4 mg in all
treated groups

ALT ¼ alanine aminotransferase; AST ¼ aspartate aminotransferase; DEE ¼ drug-eluting embolic; DOX ¼ doxorubicin; EO ¼ ethiodized oil; FdUrd-C8 ¼ 30 ,50 -dioctanoyl-5-fluoro-20 -
deoxyuridine; LHA ¼ left hepatic artery; MHA ¼ middle hepatic artery; PHA ¼ proper hepatic artery.
 Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017
188
Table 3. Dosimetric Data, Tumor Response, and Survival Data Obtained in Animal Models after Intraarterial Injection of Different Types of Re-Ethiodized Oil (19,24,28–32)

188
Study, Year Re-EO Animal Dosimetry Tumor Survival
Model Response
Injected Dose Biological Radiation Dose Tumor Uptake Tumor-to-Liver
(MBq) Half-Life (% ID/g) Ratio
Wang et al (28), EDTB-EO Rat N1-S1 7.4 122.9 h in tumor, 128.5 mGy/MBq in 1 h, 13.6; 24 h, 11.4; 1 h, 5.2; 24 h, 7.7; – –
1996 31.7 h in liver tumor; 1 mGy/ 48 h, 10.2 48 h, 10.8
MBq in liver
188
Garin et al (29), Re-SSS- EO Rat N1-S1 5.5 – – 1 h, 14.3; 24 h, 17.2; 1 h, 2.9; 24 h, 4.1; – –
2004 48 h, 18.6 48 h, 4.1
Luo et al (30), ECD- EO Rat N1-S1 7.4 – – 1 h, 11.2; 24 h, 7.3; 1 h, 13.2; 24 h, 8.5; – –
2004 48 h, 3.6 48 h, 6.8
Yoon et al (19), HDD-EO Rabbit VX2 705 16.2 h in tumor 147.7 Gy in tumor – – Tumor growth –
2004 ratio (vs pre-
treatment): –3.6
(1 wk), –7.6
(2 wk), –11.1
(3 wk); mean
0.3% viable
tumor
Tang et al (31), MN-16ET/EO Rat N1-S1 3.7 – – 1 h, 11.5; 24 h, 13.2; – – –
2011 48 h, 10.7
Huang et al (32), MN-16ET/EO Rat N1-S1 18.5 – – – – Significant > 60 d
2013 decrease in
tumor size,
whereas tumors
grew in
untreated/IA EO
alone–treated
rats
Lin et al (24), MN-16ET/EO Rat N1-S1 7.4 – – – – Tumor size > 56 d
2013 decrease of 56%
at 8 wk, 100%
responders

ECD ¼ ethyl cysteinate dimer; EDTB ¼ N,N,N0 ,N0 -tetrakis(2-benzymidazolylmethyl)-1,2-ethanediamine; EO ¼ ethiodized oil; HDD ¼ hexadecyl 2,2,9,9-tetramethyl-4,7-diaza-1,10-
decanethiol; IA ¼ intraarterial; MN-16ET ¼ N-[2-(triphenylmethyl)thioethyl]-3-aza-19-ethyloxycarbonyl-3-[2-(triphenylmethyl)thioethyl]octadecanoate; ID ¼ injected dose;
SSS ¼ Super-Six Sulfur.

9
10 ▪ Transarterial Therapies with Ethiodized Oil in Liver Cancer Models Gaba et al ▪ JVIR

10–50-fold higher than maximal therapeutic levels achieved p53/ethiodized oil versus 25.8 days for untreated animals
orally (94.2 vs 2–10 μg/mL). In 2013, Chatziioannou et al and 39.7 days for ethiodized oil alone.
(34) reported outcomes of sorafenib/ethiodized oil chemo- Zou et al (44) evaluated the effect of VEGF small inter-
embolization in 6 rabbits, describing high intratumoral sor- fering RNA (siRNA) supplemented with lipofectamine and
afenib concentrations at 24 hours (794 ng/g) and 72 hours mixed with ethiodized oil on the growth of VX2 tumors. At
(64 ng/g) after treatment associated with high liver-to-serum 28 days after treatment, tumor growth ratios were 36.1%,
drug ratios (14- and 22-fold). In 2014, Zhang et al (35) 79.7%, and 155.2% for the low-dose VEGF-siRNA group,
demonstrated that transarterial administration of ethiodized high-dose VEGF-siRNA group, and untreated group,
oil/sorafenib better inhibited tumor growth in a VX2 model respectively. Microvessel density decreased from 57.9% in
compared with other LRTs (35). In 2015, Parvinian et al the untreated group to 34.2% and 22.6% in the low-dose and
(36) studied the PKs of conventional chemoembolization high-dose groups.
with sorafenib in VX2 tumors and observed a 5:1 prefer- Image-Guided Nanoembolization.—In 2013, Mouli et al
ential tumor uptake versus normal liver, but with systemic (45) reported that nanoembolization of doxorubicin super-
washout resulting in relatively high circulating drug levels. paramagnetic iron-oxide nanoparticles emulsified in ethio-
In 2016, Kim et al (37) compared ethiodized oil emboliza- dized oil resulted in a 240%–260% increase in intratumoral
tion versus ethiodized oil/sorafenib chemoembolization in nanoparticle accumulation compared with intravenous
20 VX2 rabbits and reported high intrahepatic sorafenib administration. Jeon et al (46) described the physicochem-
concentrations, 400-fold greater than serum levels, which ical characteristics of doxorubicin-loaded porous magnetic
was associated with greater tumor necrosis compared with nanoclusters (pMNCs). After transarterial infusion of
ethiodized oil alone (91.1% vs 63.8%). In 2016, Seiden- doxorubicin-loaded pMNCs/ethiodized oil in a VX2
sticker et al (38) compared tumor response across VX2 model, MR imaging and histologic evaluations revealed
tumors treated with sorafenib/ethiodized oil chemo- that the long-term drug release and retention of
embolization versus systemic oral sorafenib and found that doxorubicin-loaded pMNCs within ethiodized oil
chemoembolization was associated with greater tumor size enhanced liver cancer cell death. Based on these suc-
reduction (3.3 vs þ8.0 mm) and devascularization (100% cesses, ethiodized oil–based nanoembolization has been
vs 50% devascularized). employed by other investigators studying nanoparticle
Delivery of Gene Therapy.—The utility of ethiodized oil LRT (47–49).
for intracellular delivery of nucleic acid through viral Immunoembolization with Lipiodol.—Immunoemboliza-
vectors or naked DNA complexes has been investigated in tion refers to transarterial administration of immunomodu-
multiple studies. Wu et al (39) explored the inhibitory ef- lators or immune effector cells aimed at inciting the immune
fect of vascular endothelial growth factor (VEGF) antisense system to kill tumor cells or directly killing tumor cells. Xia
oligodeoxynucleotides (ODNs) on VEGF expression and et al (50) demonstrated that “sandwich” technique intra-
the antitumor effect after intra-arterial infusion of antisense arterial interleukin (IL)-12 gene therapy combined with
ODNs mixed with ethiodized oil in a rat Walker-256 hep- ethiodized oil/mitomycin C chemoembolization can signif-
atoma model. Tumor growth rate was significantly lower in icantly reduce the growth of VX2 tumors, with a significant
the ODN/ethiodized oil group than in the ethiodized oil and increase in apoptotic index compared with conventional
untreated groups (140.1% vs 177.9% and 403.9%). VEGF chemoembolization or IL-12 alone. IL-12 and ethiodized oil
messenger RNA expression and microvessel density in chemoembolization resulted in a reduced tumor growth ratio
tumorous tissues were lowest in the ODN/ethiodized oil (182.7% vs 416.9%).
group (39). Intra-arterial injection of ethiodized oil/ODN- Tumor necrosis factor (TNF) has been combined with
50 -isothiocyanate in the same rat model revealed preferen- ethiodized oil to treat liver tumors in a rabbit VX2 model (51).
tial fluorescence in tumor at 1 and 3 days after treatment At 1 and 3 days after treatment, intratumoral TNF concen-
(40). trations were 7- and 10-fold higher in an ethiodized oil/TNF–
In 2006, Kim et al (41) explored intra-arterial nonviral treated group compared with a group treated with TNF alone.
(pCMV-lucþ) gene delivery of DNA/polyethylenimine Ethiodized oil/TNF treatment revealed more prominent tu-
complex ethiodized oil emulsion in a VX2 model. Twenty- mor growth inhibition compared with TNF alone at 7 days
four hours after administration, luciferase activity in tumor after treatment (approximately 60% vs less than 40%).
was significantly greater in the ethiodized oil group. In Novel Chemotherapy Agents.—The antitumor efficacy of
2007, Gu et al (42) compared the response of VX2 tumors lidamycin—a potent antibiotic agent with an in vitro activity
after transarterial adenovirus-p53 gene therapy alone or with approximately 1,000 times greater than doxorubicin—and
ethiodized oil. Ethiodized oil gene therapy resulted in a ethiodized oil has been compared with that of doxorubicin/
lower tumor growth ratio 1 week after treatment (1.75 vs ethiodized oil after intra-arterial administration in a VX2
2.35) and reduced expression rate of mutant p53 (0.66 vs model (52). Lidamycin/ethiodized oil treatment showed a
0.53). In 2012, Li et al (43) studied the therapeutic effect of more powerful tumor-inhibitory effect and lowered the
polyplex p53-loaded hydroxyapatite nanoparticles in ethio- expression levels of proliferating cell nuclear antigen
dized oil in a rabbit VX2 model. The treatment significantly (CD31 marker) and VEGF compared with doxorubicin/
increased mean survival time: 60.4 days for polyplex ethiodized oil.
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017 11

DISCUSSION against reporting negative results may overestimate the ef-

ficacy and safety of the reported transarterial techniques.
Ethiodized oil first came into medical practice as a thera- Second, heterogeneity of study design, regarding the num-
peutic agent in 1901 (53). Recognition and application of ber of tested animals and—in some studies—the lack of
ethiodized oil as a tumor-seeking drug-delivery vehicle that control groups, prevent hazard ratio calculation to support
facilitates intracellular entry of chemotherapeutic agents improved efficacy and/or safety of one specific ethiodized
prompted its widespread use in transarterial chemo- oil LRT. Third, a large number of preclinical studies were
embolization beginning in the 1980s, and demanded a performed in the rabbit VX2 model, which is known to have
formal understanding and scientific foundation for ideal intrinsic limitations. Fourth, ethical considerations generally
chemotherapy mixture preparation methodologies to pro- preclude investigation of survival as a primary endpoint in
vide maximal therapeutic benefit. The results of the present animal research, limiting translation to human oncology
review demonstrate that preclinical translational studies trials, in which survival is the most recognized efficacy
recapitulate the preferential tumor uptake of anticancer endpoint.
agent and confirm good hepatic and systemic tolerance, high In conclusion, ethiodized oil–based LRT animal studies
tumor response, and enhanced survival after ethiodized oil– show preferential tumor uptake of anticancer agent, good
based LRT. Taken together, preclinical studies support the hepatic and systemic tolerance, high tumor response, and
efficacy of conventional chemoembolization to carry and enhanced survival after ethiodized oil LRT. Innovative
locally accumulate anticancer agent within liver tumors, ethiodized oil–based LRTs can be evaluated with animal
with a demonstrated therapeutic efficacy. models, notably by comparing new therapies versus the
The pool of preclinical data compiled herein serves as a published agents reported in the present systematic review.
foundation on which new conventional chemoembolization
approaches may address unmet clinical needs. As evidenced
in the preclinical clinical literature, ethiodized oil can vec- REFERENCES
torize and deliver a broad spectrum of anticancer agents to
liver tumors. By mixing innovative anticancer agents in 1. Gaba RC, Lewandowski RJ, Hickey R, et al. Transcatheter therapy for
hepatic malignancy: standardization of terminology and reporting criteria.
ethiodized oil, conventional chemoembolization can further J Vasc Interv Radiol 2016; 27:457–473.
assume value as a therapeutic approach to target the hallmarks 2. Kan Z, Wright K, Wallace S. Ethiodized oil emulsions in hepatic micro-
of cancer in future medical practice. Even though ethiodized circulation: in vivo microscopy in animal models. Acad Radiol 1997; 4:
275–282.
oil has predominantly been employed as a drug-delivery 3. Chou FI, Fang KC, Chung C, et al. Lipiodol uptake and retention by human
agent for conventional chemotherapeutic drugs, other inno- hepatoma cells. Nucl Med Biol 1995; 22:379–386.
vative anticancer agents (radionuclide, antiangiogenic, gene, 4. Kan Z, Wallace S. Sinusoidal embolization: impact of iodized oil on hepatic
microcirculation. J Vasc Interv Radiol 1994; 5:881–886.
nanoembolic, and immune therapies) have been validated in 5. Lencioni R, de Baere T, Soulen MC, Rilling WS, Geschwind JF. Lipiodol
animal models. Combining ethiodized oil with these new transarterial chemoembolization for hepatocellular carcinoma: A system-
therapies for clinical translation should allow enhancement of atic review of efficacy and safety data. Hepatology 2016; 64:106–116.
6. de Baere T, Dufaux J, Roche A, et al. Circulatory alterations induced by
conventional chemoembolization efficacy, moving toward a intra-arterial injection of iodized oil and emulsions of iodized oil and
“conventional chemoembolization 2.0” technique. doxorubicin: experimental study. Radiology 1995; 194:165–170.
Preclinical animal models and translational investigations 7. Kan Z, Ivancev K, Hagerstrand I, Chuang VP, Lunderquist A. In vivo
microscopy of the liver after injection of Lipiodol into the hepatic artery
represent pivotal tools to develop novel LRTs. Although and portal vein in the rat. Acta Radiol 1989; 30:419–425.
human clinical trials are the benchmark for advancing 8. Kan Z, Sato M, Ivancev K, et al. Distribution and effect of iodized pop-
standard-of-care cancer therapeutic agents, regulatory and pyseed oil in the liver after hepatic artery embolization: experimental
study in several animal species. Radiology 1993; 186:861–866.
financial burdens of opening trials are significant and time- 9. Kan Z. Dynamic study of iodized oil in the liver and blood supply to hepatic
consuming, and patient enrollment is challenging as a result tumors. An experimental investigation in several animal species. Acta
of stringent eligibility criteria and competing clinical trials. Radiol Suppl 1996; 408:1–25.
10. Kan Z, Ivancev K, Lunderquist A, et al. In vivo microscopy of hepatic
Studies in preclinical animal models are therefore critical in tumors in animal models: a dynamic investigation of blood supply to
that they can efficiently and effectively undergo cohort hepatic metastases. Radiology 1993; 187:621–626.
clinical trial participation, successfully eliminating accrual 11. de Baere T, Zhang X, Aubert B, et al. Quantification of tumor uptake of
iodized oils and emulsions of iodized oils: experimental study. Radiology
and logistical barriers to permit prospective early-phase 1996; 201:731–735.
assessment of therapeutic modalities and to establish the 12. Choi JW, Cho HJ, Park JH, et al. Comparison of drug release and phar-
validity of new technologies. macokinetics after transarterial chemoembolization using diverse Lipiodol
emulsions and drug-eluting beads. PloS One 2014; 9:e115898.
The present systematic review has limitations. First, the 13. Gaba RC, Baumgarten S, Omene BO, et al. Ethiodized oil uptake does not
heterogeneity of several parameters, including the type of predict doxorubicin drug delivery after chemoembolization in VX2 liver
ethiodized oil emulsion, nature and dose of chemothera- tumors. J Vasc Interv Radiol 2012; 23:265–273.
14. Yin X, Guo Y, Li W, et al. Chemical shift MR imaging methods for the
peutic drugs, and type of animal model, prevents identifi- quantification of transcatheter lipiodol delivery to the liver: preclinical
cation of quantitative reference data for ethiodized oil–based feasibility studies in a rodent model. Radiology 2012; 263:714–722.
LRT. Moreover, LRTs employing ethiodized oil suspensions 15. Fukushima S, Kawaguchi T, Nishida M, et al. Selective anticancer effects
of 3’,5’-dioctanoyl-5-fluoro-2’-deoxyuridine, a lipophilic prodrug of 5-fluoro-
were excluded even though efficacy has been demonstrated 2’-deoxyuridine, dissolved in an oily lymphographic agent on hepatic
in several nonclinical studies. Additionally, publication bias cancer of rabbits bearing VX-2 tumor. Cancer Res 1987; 47:1930–1934.
12 ▪ Transarterial Therapies with Ethiodized Oil in Liver Cancer Models Gaba et al ▪ JVIR

16. Kawaguchi T, Fukushima S, Hayashi Y, Kaneko M, Nakano M. Selective growth and intrahepatic metastasis in rabbit VX2 hepatocellular carcinoma
accumulation of 3’,5’-dioctanoyl-5-fluoro-2’-deoxyuridine (FdUrd-C8) and model. Int J Clin Exp Pathol 2014; 7:7775–7781.
sustained release of its active metabolites in VX-2 rabbit hepatoma 36. Parvinian A, Casadaban LC, Hauck ZZ, van Breemen RB, Gaba RC.
following intraarterial administration of FdUrd-C8 solution in Lipiodol. Pharmacokinetic study of conventional sorafenib chemoembolization in a
Cancer Res 1988; 48:4179–4183. rabbit VX2 liver tumor model. Diagn Interv Radiol 2015; 21:235–240.
17. Konno T. Targeting cancer chemotherapeutic agents by use of lipiodol 37. Kim GM, Kim MD, Kim do Y, et al. Transarterial chemoembolization using
contrast medium. Cancer 1990; 66:1897–1903. sorafenib in a rabbit VX2 liver tumor model: pharmacokinetics and anti-
18. Yoon CJ, Chung JW, Park JH, et al. Transcatheter arterial chemo- tumor effect. J Vasc Interv Radiol 2016; 27:1086–1092.
embolization with paclitaxel-lipiodol solution in rabbit VX2 liver tumor. 38. Seidensticker M, Streit S, Nass N, et al. Modified transarterial chemo-
Radiology 2003; 229:126–131. embolization with locoregional administration of sorafenib for treating
19. Yoon CJ, Chung JW, Park JH, et al. Transcatheter arterial embolization hepatocellular carcinoma: feasibility, efficacy, and safety in the VX-2 rabbit
with 188Rhenium-HDD-labeled iodized oil in rabbit VX2 liver tumor. J Vasc liver tumor model. Diagn Interv Radiol 2016; 22:378–384.
Interv Radiol 2004; 15:1121–1128. 39. Wu HP, Feng GS, Liang HM, Zheng CS, Li X. Vascular endothelial growth
20. Baumgarten S, Gaba RC, van Breemen RB. Confirmation of drug delivery factor antisense oligodeoxynucleotides with lipiodol in arterial emboliza-
after liver chemoembolization: direct tissue doxorubicin measurement by tion of liver cancer in rats. World J Gastroenterol 2004; 10:813–818.
UHPLC-MS-MS. Biomed Chromatogr 2012; 26:1529–1533. 40. Wu HP, Feng GS, Tian Y. Hepatic artery infusion of antisense oligo-
21. Becker S, Lepareur N, Cadeillan V, et al. Optimization of hepatocarcinoma deoxynucleotide and Lipiodol mixture transfect liver cancer in rats. World
uptake with radiolabeled lipiodol: development of new Lipiodol formula- J Gastroenterol 2005; 11:2408–2412.
tions with increased viscosity. Cancer Biother Radiopharm 2012; 27: 41. Kim YI, Chung JW, Park JH, Han JK, Hong JW, Chung H. Intraarterial
149–155. gene delivery in rabbit hepatic tumors: transfection with nonviral vector by
22. Ueda T, Murata S, Mine T, et al. Comparison of epirubicin-iodized oil using iodized oil emulsion. Radiology 2006; 240:771–777.
suspension and emulsion for transcatheter arterial chemoembolization in 42. Gu T, Li CX, Feng Y, Wang Q, Li CH, Li CF. Trans-arterial gene therapy for
VX2 tumor. Sci World J 2012; 2012:961–986. hepatocellular carcinoma in a rabbit model. World J Gastroenterol 2007;
23. Liang B, Xiong F, Wu H, et al. Effect of transcatheter intraarterial therapies 13:2113–2117.
on the distribution of doxorubicin in liver cancer in a rabbit model. PloS 43. Li G, Ye L, Pan J, et al. Antitumoural hydroxyapatite nanoparticles-
One 2013; 8:e76388. mediated hepatoma-targeted trans-arterial embolization gene therapy:
24. Lin WY, Luo TY, Tsai SC, Kao CH, Tang IC, Huang PW. A comparison of in vitro and in vivo studies. Liver Int 2012; 32:998–1007.
Re-188-MN-16ET-lipiodol and transcatheter arterial chemoembolization in 44. Zou Y, Guo CG, Yang ZG, Sun JH, Zhang MM, Fu CY. A small interfering
the treatment of hepatoma: an animal study. Nucl Med Biol 2013; 40: RNA targeting vascular endothelial growth factor efficiently inhibits
437–441. growth of VX2 cells and VX2 tumor model of hepatocellular carcinoma in
25. Tomozawa Y, Nitta N, Ohta S, et al. Antitumor effect of miriplatin-Lipiodol rabbit by transarterial embolization-mediated siRNA delivery. Drug Des
suspension/emulsion using a VX2 liver tumor model. Jpn J Radiol 2013; Devel Ther 2016; 10:1243–1255.
31:662–667. 45. Mouli SK, Tyler P, McDevitt JL, et al. Image-guided local delivery strate-
26. Lilienberg E, Ebeling Barbier C, Nyman R, et al. Investigation of hep- gies enhance therapeutic nanoparticle uptake in solid tumors. ACS Nano
atobiliary disposition of doxorubicin following intrahepatic delivery of 2013; 7:7724–7733.
different dosage forms. Mol Pharm 2014; 11:131–144. 46. Jeon MJ, Gordon AC, Larson AC, Chung JW, Kim YI, Kim DH. Trans-
27. Akashi Y, Koreeda C, Mizuno T, Inoue K, Kawa SK, Tanaka Y. Hepatic catheter intra-arterial infusion of doxorubicin loaded porous magnetic
parenchymal changes after the intraarterial injection of lipiodol in tumor- nano-clusters with iodinated oil for the treatment of liver cancer. Bio-
bearing rabbits. Invest Radiol 1993; 28:128–132. materials 2016; 88:25–33.
28. Wang SJ, Lin WY, Chen MN, et al. Biodistribution of rhenium-188 Lipiodol 47. Yin Q, Yap FY, Yin L, et al. Poly(iohexol) nanoparticles as contrast agents
infused via the hepatic artery of rats with hepatic tumours. Eur J Nucl for in vivo X-ray computed tomography imaging. J Am Chem Soc 2013;
Med 1996; 23:13–17. 135:13620–13623.
29. Garin E, Denizot B, Noiret N, et al. 188Re-SSS lipiodol: radiolabelling and 48. Dong S, Tang Q, Long M, Guan J, Ye L, Li G. The cooperative effect of
biodistribution following injection into the hepatic artery of rats bearing p53 and Rb in local nanotherapy in a rabbit VX2 model of hepatocellular
hepatoma. Nucl Med Commun 2004; 25:1007–1013. carcinoma. Int J Nanomedicine 2013; 8:3757–3768.
30. Luo TY, Hsieh BT, Wang SJ, et al. Preparation and biodistribution of 49. Tam AL, Melancon MP, Abdelsalam M, et al. Imaging intratumoral
rhenium-188 ECD/Lipiodol in rats following hepatic arterial injection. Nucl nanoparticle uptake after combining nanoembolization with various abla-
Med Biol 2004; 31:671–677. tive therapies in hepatic VX2 rabbit tumors. J Biomed Nanotechnol 2016;
31. Tang IC, Luo TY, Liu SW, et al. Synthesis and application of 188Re-MN- 12:296–307.
16ET/Lipiodol in a hepatocellular carcinoma animal model. Nucl Med Biol 50. Xia X, Li X, Feng G, Zheng C, Liang H, Zhou G. Intra-arterial interleukin-12
2011; 38:1043–1052. gene delivery combined with chemoembolization: anti-tumor effect in a
32. Huang PW, Tsai SC, Luo TY, Kao CH, Lin WY. Therapeutic efficacy of rabbit hepatocellular carcinoma (HCC) model. Acta Radiol 2013; 54:
188Re-MN-16ET lipiodol in an animal model of hepatocellular carcinoma. 684–689.
Ann Nucl Med 2013; 27:532–537. 51. Watanabe D, Ueo H, Inoue H, et al. Antitumor effects of intraarterial
33. Gaba RC, Yap FY, Martinez EM, et al. Transarterial sorafenib chemo- infusion of tumor necrosis factor/Lipiodol emulsion on hepatic tumor in
embolization: preliminary study of technical feasibility in a rabbit model. rabbits. Oncology 1995; 52:76–81.
J Vasc Interv Radiol 2013; 24:744–750. 52. Zhong G, Qi J, Huo S, et al. Transarterial oily chemoembolization with
34. Chatziioannou AN, Siskos AP, Loxas D, et al. Transarterial embolization lidamycin shows potent therapeutic efficacy in VX2 rabbit liver tumor.
with sorafenib in animal livers: a pharmacokinetics study. J Vasc Interv Oncol Targets Ther 2015; 8:3079–3086.
Radiol 2013; 24:1657–1663. 53. Bonnemain B, Guerbet M. The history of Lipiodol (1901-1994) or how a
35. Zhang L, Liu FY, Fu JX, Duan F, Fan QS, Wang MQ. Hepatic arterial medication may evolve with the times. Rev Hist Pharm (Paris) 1995; 42:
administration of sorafenib and iodized oil effectively attenuates tumor 159–170.