Safe Lipid Nanocapsule-based Gel technology to Target Lymph Nodes and Combat
Mediastinal Metastases from an Orthotopic Non-Small-Cell Lung Cancer model in
SCID-CB17 mice
Nathalie WAUTHOZ 1,2, Guillaume BASTIAT 1,2,*, Elodie MOYSAN 1,2, Anna CIESLAK 1,2,
Kazuya KONDO 3, Marc ZANDECKI 4, Valérie MOAL 5, Marie-Christine ROUSSELET 6, José
HUREAUX 1,7 and Jean-Pierre BENOIT 1,2
Word count for the abstract: 143
Word count for the manuscript (including body text, and figure and table legends):
5206
Number of references: 42
Number of figures: 3
Number of tables: 3
1
Safe Lipid Nanocapsule-based Gel technology to Target Lymph Nodes and Combat
Mediastinal Metastases from an Orthotopic Non-Small-Cell Lung Cancer model in
SCID-CB17 mice
Nathalie WAUTHOZ 1,2, Guillaume BASTIAT 1,2,*, Elodie MOYSAN 1,2, Anna CIESLAK 1,2,
Kazuya KONDO 3, Marc ZANDECKI 4, Valérie MOAL 5, Marie-Christine ROUSSELET 6, José
HUREAUX 1,7 and Jean-Pierre BENOIT 1,2
1
2
LUNAM Université – Micro et Nanomédecines Biomimétiques, F-49933 Angers, France
INSERM – U1066 IBS-CHU, F-49933 Angers, France; nathalie.wauthoz@univ-angers.fr,
guillaume.bastiat@univ-angers.fr, elodie.moysan@etud.univ-angers.fr, anna.cieslak@etud.univangers.fr, jean-pierre.benoit@univ-angers.fr
3
Department of Oncological Medical Services, Institute of Health Biosciences, The University of
Tokushima Graduate School, 18-15 Kuramoto-cho 3 Tokushima, 770-8509, Japan;
kondo@clin.med.tokushima-u.ac.jp
4
Hematology Department, Academic Hospital, Angers, F-49933, France; MaZandecki@chuangers.fr
5
Biochemistry Department, Academic Hospital, Angers, F-49933, France; VaMoal@chuangers.fr
6
Cell and Tissue Pathology Department, Academic Hospital, Angers, F-49933, France;
MCRousselet@chu-angers.fr
7
Pneumology Department, Academic Hospital, Angers, F-49933, France; johureaux@chuangers.fr
2
* To whom correspondence should be addressed:
Tel: +33 244688531, Fax: +33 244688546, E-mail: guillaume.bastiat@univ-angers.fr
The authors declare that there are no conflicts of interest.
Funding sources: this work has been carried out within the research program LYMPHOTARG
financially supported by EuroNanoMed ERA-NET 09 and by the Région Pays de la Loire.
Abstract
The purpose of this study is the assessment of gel technology based on a lauroyl derivative of
gemcitabine encapsulated in lipid nanocapsules delivered subcutaneously or intravenously after
dilution to (i) target lymph nodes, (ii) induce less systemic toxicity and (iii) combat mediastinal
metastases from an orthotopic model of human, squamous, non-small-cell lung cancer Ma44-3
cells implanted in severe combined immunodeficiency mice.
The gel technology mainly targeted lymph nodes as revealed by the biodistribution study.
Moreover, the gel technology induced no significant myelosuppression (platelet count) in
comparison with the control saline group, unlike the conventional intravenous gemcitabine
hydrochloride treated group (p < 0.05). Besides, the gel technology, delivered subcutaneously
twice a week, was able to combat locally mediastinal metastases from the orthotopic lung tumor
and to significantly delay death (p < 0.05) as was the diluted gel technology delivered
intravenously three times a week.
Keywords: lymphatic targeting, nanomedicine, mediastinum, immunomodulation
3
Background
Lung cancer remains the leading cause of cancer-related mortality around the world [1, 2]. Nonsmall-cell lung cancers (NSCLC) represent 85% of all cases of lung cancers [3]. Diagnosed
NSCLC are mainly present in regional and advanced extent for 22% and 56% of patients,
respectively [4]. At these stages, cells from a primary tumor (i.e. metastases) have migrated and
have been disseminated by blood and/or lymphatic vessels to distal organs. Via the lymphatic
route, metastases are trapped by lymph nodes (firstly by sentinel nodes followed by secondary
and distal nodes) [5]. The more distal the invasion of lymph nodes is, the higher the malignancy
and the poorer the prognosis is [5]. Indeed, the five-year survival rate of patients with NSCLC
treated by current treatment methods is estimated at 42% for a stage of N0 (without regional
lymph node metastases) to 7% for N3 corresponding to the latest stage of lymph node implication
[6]. Patients with mediastinum lymph node metastases (N2 and N3) are considered in the
advanced stage III and are mainly treated by a combination of radiotherapy and chemotherapy
with poor improvement in long-term survival and at the expense of a large rise in Grade 3
toxicities [7]. Moreover, mediastinal lymph node metastases can induce the “superior vena cava
(SVC) syndrome”, which is an array of clinical signs and symptoms due to the impairment of
blood flow through the SVC [8, 9]. The flow impairment is usually caused by the enlargement of
mediastinal lymph nodes that compress the SVC. The symptoms include dyspnea, coughing, and
swelling of the face, neck, upper trunk, and extremities. Furthermore, headache, confusion, and
possibly coma can result from cerebral edema. The SVC syndrome from lung malignancies is
treated by chemotherapy and/or radiotherapy or by endovascular stenting in emergency cases [8,
9].
4
Conventional chemotherapy is mainly delivered by perfusion but includes major limitations: high
systemic toxicity and poor lymph node exposure and retention [10]. In consequence, an efficient
drug delivery system increasing lymph node exposure and retention while decreasing systemic
toxicity could be useful against lymph-metastatic cancers [5]. With this aim, a nanocarrier
system, lipid nanocapsules (LNCs), loaded with the lipophilic pro-drug gemcitabine (Gem-C12)
has been developed. Gemcitabine is a third generation agent composed of the platinum based
doublet chemotherapy used as first-line treatment for advanced NSCLC patients with good
performance status [3]. However, due to its hydrophilic nature and its low membrane
permeability and also due to its extensive deamination in plasma and tissues by cytidine
deaminase, its plasma elimination half-life (t1/2) after 30 min. perfusion is extremely short (42-94
min.) [11]. Therefore, a high dose regimen is required which mainly leads to dose-limiting
myelosuppression in approximately two-thirds of patients [12]. Moreover, different resistance
mechanisms against gemcitabine have evolved in cancer cells; this fostered the emergence of
chemically-modified gemcitabine [11]. The Gem-C12 (i.e. 4-(N)-lauroyl-gemcitabine) was
synthetized by Tokunaga et al to decrease the deamination in blood responsible for inactive
metabolite formation and to decrease the hydrophilicity responsible for poor drug loading in
lipophilic nanosystems [11].
The LNC loaded with Gem-C12 has previously demonstrated the ability (i) to form a hydrogel
when Gem-C12 is encapsulated in LNC, (ii) to release Gem-C12-based LNC at a rate depending
on the Gem-C12 concentration, (iii) to be in vitro injected through 18 or 21G needle, (iv) to form
a suspension after dilution, and (v) to exert a higher-than-average in vitro anticancer activity
against human NSCLC and prostatic carcinoma cell lines [13]. For this study, a subcutaneous (sc)
administration route and a 50nm LNC with a polyethyleneglycol 660-based surface forming the
5
hydrogel were chosen because they seemed to be the most appropriate for lymphatic drainage
and lymph node retention [5, 14-19]. Indeed, after sc administration, nanocarriers whose size was
less than 0.1µm have direct access to lymphatic capillaries; absorption by blood capillaries is
restricted to water and small hydrophilic molecules (< 10nm) [5, 17, 19]. Moreover, having no
overall charges (or negative charges) generally helps gain access to the lymphatic system, but
hydrophobic (macrophage phagocytosis) or increased size is required for lymph node retention.
Indeed, mechanical depth filtration in the meshwork of reticular cells and phagocytosis by
macrophages are generally accepted as the major mechanisms of liposome uptake in lymph nodes
[5, 17].
Therefore, the purpose of this study is to evaluate in vivo this LNC-based gel technology in terms
of (i) lymph node targeting, (ii) tolerance, and (iii) antitumor efficacy in a highly aggressive
patient-like lymphogenous metastatic model of human NSCLC. This lymphogenous metastatic
model mimics the spreading of metastases in mediastinum from the primary tumor implanted in
the lung of severe combined immunodeficiency mice with a CB17 genetic background (SCIDCB17) as observed for NSCLC patients [20].
Methods
Material
Labrafac® WL 1349 (caprylic–capric acid triglycerides) was provided by Gattefossé S.A. (SaintPriest, France). Kolliphor® HS15 (mixture of free polyethyleneglycol 660 and polyethyleneglycol
660 hydroxystearate) was supplied by BASF (Ludwigshafen, Germany). Sodium chloride,
acetone and ethanol were purchased from VWR (Fontenay-sous-Bois, France). Span® 80
(sorbitane monooleate), Tween® 80 (polysorbate 80) and 5-aminolevulinic acid (ALA) were
6
purchased
from
Sigma
(St
Quentin
Fallavier,
France).
1,1'-Dioctadecyl-3,3,3',3'-
tetramethylindodicarbocyanine (DiD) was provided by Life Technologies (Saint-Aubin, France).
Methanol was analytical grade purchased from Fischer Scientific (Loughborough, United
Kingdom). Water was obtained from a MilliQ filtration system (Millipore, Paris, France). 4-(N)lauroyl-gemcitabine (Gem-C12) was synthesized and characterized as described elsewhere [13].
Preparation of the formulations
The LNC-based gel technology was prepared following the phase-inversion temperature (PIT)
process [13, 21]. Briefly, Gem-C12 (0.062g) was first solubilized (5%, ratio Gem-C12/Labrafac®
w/w) in a mix of Labrafac® WL 1349 (1.24g), Span® 80 (0.25g) and acetone that was
evaporated before the addition of Kolliphor ® HS15 (0.967g), NaCl (0.045g) and water
(1.02g). All were then mixed and heated to 75°C under magnetic stirring at 500rpm followed by
cooling to 45°C (rate of 5°C/min). Three cycles were performed and at the last cooling phase, a
sudden dilution with 2.12g room-temperature water was carried out at 60°C. LNC loaded with
Gem-C12 spontaneously formed a hydrogel with a waxy aspect. The gelation process was
considered to be achieved after 24h at 4°C. LNC loaded with the prodrug (Gem-C12 LNC) and
used in a liquid form were acquired by the 4-fold dilution of the formulation directly after the
process, before the establishment of the hydrogel. Non-loaded LNC was prepared in the same
way as Gem-C12 LNC hydrogel but without Gem-C12 and acetone. Fluorescent LNC was
obtained by adding the fluorescent probe (DiD) with the other reagents at a final concentration of
0.1% of Labrafac (w/w).
Free Gem-C12 was dissolved in ethanol, Tween® 80 and water
(87.6/5.5/6.9 v/v/v) as a micellar system for in vivo experiments. The commercial gemcitabine
7
hydrochloride used was Gemcitabine Hospira 38mg/mL, diluted at the required concentration
with 0.9%. NaCl.
Characterization of LNC-based formulations
The hydrodynamic diameter (Z-average), polydispersity index (PdI) and zeta potential of LNCbased formulations were determined by dynamic light scattering using a Zetasizer® Nano series
DTS 1060 (Malvern Instruments S.A., Worcestershire, UK). LNC-based formulations were
diluted 1:60 (v/v) in deionised water in order to ensure a convenient scatter intensity on the
detector; each measurement was done in triplicate at 25°C. The drug loading was determined
after dialysis, using an Ultra Performance Liquid Chromatography apparatus with UV
spectroscopy detection as previously described [13]; each measurement was done in triplicate.
Animals
For pharmacokinetic and biodistribution studies, female nude Swiss mice were purchased from
Harlan (Gannat, France) and for in vivo efficacy and tolerance evaluations, male SCID-CB17
mice were purchased from Charles River (L’Arbresle, France). The animals are housed and
maintained at the University animal facility (SCAHU) in disposable plastic cages with hardwood
chips bedding in an air-conditioned room with a 12-hour-light/12-hour-dark cycle. All the animal
experiments were carried out in accordance with EU Directive 2010/63/EU and with the
agreement of the “Comité d’Ethique pour l’Expérimentation Animale des Pays de la Loire”
(authorization CEEA; 2012-37 and 2012-73).
8
Pharmacokinetic and biodistribution of fluorescent LNC-based formulations in healthy
nude mice
Mice (9 weeks old, n = 5 for each group) were anesthetized (isoflurane) and either 110µL of
DiD-Gem-C12 LNC was injected subcutaneously (sc) behind the neck, or 110µL of DiD LNC
was injected intravenously (iv) into the tail vein with the aim of delivering the same amount of
DiD-loaded LNC. After various lengths of time, from 1h to 336h, the mice were sacrificed and
the blood was removed by cardiac puncture in a venous blood collection tube containing Liheparin (Tube Micro from SARSTEDT, Marnay, France). The organs were then removed for the
biodistribution study. Plasma from each blood sample was obtained after 10 min. of
centrifugation at 2,000g.
The DiD concentrations in plasma (encapsulated inside LNC) over time were determined using a
microplate reader Fluoroscan Ascent® (Labsystems SA, Cergy-Pontoise, France) at excitation
and emission wavelengths of 646 and 678nm, respectively. The DiD concentrations in plasma
were intrapolated from the linear curve between 0.006 and 12µg/mL (r2 > 0.999). The recovered
DiD concentrations were then normalized in function of the animal’s weight, assuming that blood
represents 7.5% of mouse body weight [22]. Pharmacokinetic data were determined by iv bolus
and extravascular non-compartmental analysis (for iv and sc administration, respectively) of the
recovered systemic DiD concentration over time using Kinetica 4.1.1 software (Thermo Fisher
Scientific, Villebon sur Yvette, France). The trapezoidal calculation method was used to
determine the area under the curve (AUC) during the whole experimental period (from 1 to 336h)
without extrapolation. The t1/2 was calculated from 1 to 8h for t1/2 distribution and from 8 to 336h
for t1/2 elimination.
9
For the biodistribution study, the organs (i.e., kidneys, liver, spleen, lung, heart, stomach, and
intestine), and lymph nodes (i.e., inguinal, axillary, cervical and brachial lymph nodes) were
removed and analyzed by a fluorescence CRI MaestroTM imaging system (Woburn, USA). Semiquantitative data were obtained by setting a time exposure of 10ms between 630 and 800nm,
unmixing the generated cube, extracting the background, and drawing the regions of interest from
fluorescence images (Figure 1-A). The software Maestro 2.10 (Woburn, USA) was used to
calculate the average signal expressed in photon/cm²/s.
Cell culture and inoculum preparation for lung implantation
An Ma44-3 cell line derived from the human squamous NSCLC carcinoma cancer cell line Ma44
by limit dilution method was kindly provided by Prof. Kondo (Department of Oncological and
Regenerative Surgery, School of Medicine, University of Tokushima, Japan) and was cultured in
RPMI 1640 (Lonza, Verviers, Belgium) supplemented with 10% heat-inactivated fetal bovine
serum (Lonza, Verviers, Belgium), 100U/mL of penicillin, 100µg/mL of streptomycin and
0.250μg/mL of amphotericin B from Sigma (St Quentin Fallavier, France) at 37°C in a
humidified incubator with 5% CO2. For lung implantation, the cells were harvested at 70-80%
confluence using trypsin that was inactivated by the culture medium. Then, the cell suspension
was centrifuged at 1,400rpm for 5 min. and the supernatant was removed and replaced by RPMI
1640 containing 0.1% bovine serum albumin fraction V pH 7.0 (PAA GmbH, Pasching, Austria)
which was then mixed with Matrigel® (DB Biosciences, Bedford, MA) to obtain an inoculum of
2.0 × 106 tumor cells/mL with 400µg/mL of Matrigel®. The inoculum was kept on ice for a
maximum of 2h.
10
Orthotopic intrapulmonary implantation procedure
The orthotopic intrapulmonary implantation procedure was performed as previously reported
[20]. Briefly, the 6 week-old SCID-CB17 mice were maintained in the right lateral decubitus and
anesthetized by isoflurane inhalation. A 1cm transverse incision was made on the left lateral skin
just below the inferior border of the scapula of the SCID-CB17 mice. The muscles were
separated from the ribs by sharp dissection, leaving the intercostal muscles visible. 10μL of cell
suspension (2.0 × 104 cells) were inoculated with a 30-gauge needle to a depth of about 3-5mm
into the lung through the intercostal muscles and after, the needle was promptly pulled out. Mice
were maintained in the right lateral decubitus position after injection and the skin incision was
closed with 3-0 silk (Ethicon, St-Stevens-Woluwe, Belgium). The mice were observed until
complete recovery.
In vivo efficacy on the lymphogenous metastatic model
The in vivo efficacy of different treatments on survival was evaluated on a lymphogenous
metastatic model. Seven groups of 10 SCID-CB17 mice implanted with 2.0 × 104 Ma44-3 cells in
the left lung were randomly established. Five groups were treated by iv (145µL at each injection
by the tail vein) on Days 5, 7 and 9 after tumor grafting, either with (i) NaCl 0.9% solution
(saline iv), (ii) non-loaded LNC (non-loaded LNC iv), (iii) commercial gemcitabine
hydrochloride (Gemcitabine iv), (iv) Gem-C12 micelles (Gem-C12 iv) or (v) liquid form of GemC12-loaded LNC (Gem-C12 LNC iv). Two groups were treated by sc route (55µL at each
injection behind the neck) on Days 5 and 9 after tumor grafting either with (i) non-loaded LNC
(non-loaded LNC sc) or (ii) gel form of Gem-C12-loaded LNC (Gem-C12 LNC sc). The total
LNC delivered dose of non-loaded LNC was the same as the Gem-C12-loaded LNC for iv or sc.
11
In groups treated by gemcitabine hydrochloride or Gem-C12 (in LNC or not), the total dose
delivered to mice was 40mg (molar equivalent gemcitabine hydrochloride) per kilogram of body
weight. The mice were observed every day and body weight measurements were performed daily
during the treatment period and then three times a week. By applying the criteria for euthanasia
of experimental animals, the mice were sacrificed when they lost 20% of body weight and/or
associated with a high degree of respiratory depression. Survival analyses were carried out by
means of Kaplan-Meier curves and the log-rank test.
Visualization of the fluorescent LNC-based formulations in lung and mediastinum in
tumor bearing mice
Visualization of the lung and mediastinal distribution of fluorescent LNC-based formulations
delivered sc or iv after dilution was performed after 4h on tumor-grafted SCID-CB17 mice. On
Day 5 post tumor grafting, a group of 3 mice received an iv injection of DiD LNC and the other
group of 3 mice received an sc injection of DiD-Gem-C12 LNC to deliver the same amount of
DiD LNC. Moreover, both groups received orally 400mg/kg of ALA in PBS (200µL) to detect
tumor mass. Four hours after the respective administrations, the mice were killed and dissected
for imaging the lung and the mediastinum area. Fluorescence imaging was carried out using the
CRI Maestro system between 500 and 635nm (for Pp IX for tumor mass visualization) and 630
and 800nm (for DiD for LNC visualization) using automatic exposure.
In vivo tolerance evaluation of the formulations
Mice (7 weeks old, 5 per group) received 145µL of an iv injection of saline (saline iv),
gemcitabine hydrochloride (Gemcitabine iv), Gem-C12 micelles (Gem-C12 iv), non-loaded LNC
12
(non-loaded LNC iv) or liquid form of Gem-C12-loaded LNC (Gem-C12 LNC iv) dispersions
three times a week for one week (Monday, Wednesday and Friday). Ten mice (5 per group)
received 55µL of an sc injection of non-loaded LNC (non-loaded LNC sc) or gel form of GemC12-loaded LNC (Gem-C12 LNC sc) twice a week for one week (Monday and Friday). The LNC
delivered dose of non-loaded LNC was the same as Gem-C12-loaded LNC for iv or sc
administrations. In groups treated by gemcitabine hydrochloride or Gem-C12 (in LNC or in
micelles), the dose delivered to mice was 40mg (molar equivalent gemcitabine hydrochloride) per
kilogram of body weight. Hematology and biochemical assays on blood were carried out for each
mouse 8 days after the first injection, to assess the tolerance to treatment. Blood sampling was
performed by cardiac puncture in anaesthetized mice; half of the blood sample was placed in a
venous blood collection ethylene-diamine-tetraacetic acid tube (K2E tube from BD Microtainer,
NJ, USA) for hematology studies and the other half in a venous blood collection tube containing
heparin lithium (Tube Micro from SARSTEDT, Marnay, France) which were then centrifuged at
10,000g for 10 min. to remove the plasma and evaluate the blood biochemical markers. The
hematological parameters were determined in the Hematology Ward of the Academic Hospital of
Angers with an XE-2100 hematology analyzer (Sysmex). Plasma biochemistry analyses were
carried out at the Biochemistry Ward of the Academic Hospital of Angers on an Architect
C16000 (Abbott). Statistical comparisons of data were analyzed by a non-parametric one-way
analysis of variance (Kruskal-Wallis test), followed by Dunn's post-hoc test for pairwise
comparisons.
Results
LNC-based formulations
13
The physicochemical characteristics of LNC-based formulations are presented in Table 1. With
the addition of Gem-C12 into LNC, a hydrogel was formed that can be injected using a syringe
[13]. Once the gel was diluted 4-fold in water, a liquid LNC suspension was obtained (Gem-C12
LNC iv). LNC formulations performed by the PIT method displayed a Z-average range (i.e.,
hydrodynamic diameter) between 53 and 68nm, depending on the presence of the amphiphilic
Gem-C12 and/or DiD in the LNC surface, and a polydispersity index in all cases less than 0.1,
which means that monomodal and monodispersed distributions were obtained. In all cases, the
zeta potential was slightly negative, from -5 to -10mV. Similarly to many hydrophobic drugs [23,
24], Gem-C12 was well-encapsulated in the LNC with a high encapsulation efficiency of about
100%.
Biodistribution and pharmacokinetic of LNC-based formulations
LNC loaded with the fluorescent dye DiD were used to evaluate their tissue biodistribution and
pharmacokinetics after iv (suspension of DiD LNC) and sc (hydrogel of DiD-Gem-C12 LNC)
administration in nude Swiss mice. The fluorescent dye DiD was chosen to be encapsulated in the
LNC because (i) it is a near-infrared fluorophore limiting the auto-fluorescence wavelength
emitted by animals [25-27], (ii) it is weakly fluorescent in aqueous media, but highly fluorescent
in a lipophilic vehicle [28], and (iii) there is no dye release (strong fluorescence labeling stability
of the nanocarrier) [29]. The semi-quantitative fluorescence of the organs extracted after 1, 4, 8,
48, 96, and 336h is reported in Figure 1-C to F and Figure S1 in supplementary information. DiD
LNC iv and DiD-Gem-C12 LNC sc was rapidly distributed in the lymph nodes of the nude mice.
An immediate accumulation in the liver and spleen after 4 and 8h was observed with DiD LNC
iv. After 48h, a decrease of the accumulation was observed for these organs to totally disappear
14
after 336h (Figure 1-C). DiD-Gem-C12 LNC sc accumulated exclusively in the lymph nodes
close to the injection site (i.e., in axillary, cervical and brachial lymph nodes) in similar amounts
to the first hours (i.e., 1, 4 and 8h) as DiD LNC iv but was much higher after a few days (i.e., 48,
96 and 336h) (Figure 1-F). Moreover, DiD-Gem-C12 LNC sc presented a much lower plasmatic
exposition than DiD LNC iv (Figure 1-B) resulting in a lower AUC1-336h of 4µg.h/mL in
comparison with an AUC1-336h of 174µg.h/mL, respectively. Besides, DiD-LNC was more rapidly
cleared from the systemic circulation after iv administration with a t1/2 elimination of 19h in
comparison to 32h found for the sc injection of the hydrogel. Indeed, the latter presented
controlled-release properties [13].
In vivo efficacy of the LNC-based formulations
The patient-like lymphogenous metastatic Ma44-3 preclinical model was used to evaluate the
efficacy of LNC-based gel technology. Five days post-tumor graft in a left lung lobe, when
micrometastatic foci were detected in mediastinum (Figure S2-A), a comparative efficacy study
on randomized groups was performed with different treatments and schedules of administration.
The dose of 40mg (molar equivalent dose of gemcitabine hydrochloride) per kilogram of body
weight delivered twice or three times a week by the sc and the iv routes, respectively, was chosen
because no weight loss nor death was observed in a group of three mice using Gem-C12 LNC
(gel form) for sc route or Gem-C12 LNC (liquid form) for iv route, respectively. Higher doses or
similar doses with fewer injections caused drastic weight loss or the death of mice (data not
shown). The survival times of experimental animals were plotted on the Kaplan Meier curves as
shown in Figure 2-A. Mice of the saline iv and non-loaded LNC iv or sc groups were
characterized by a very short lifespan after tumor implantation without significant difference (p >
15
0.05, log-rank test) (Table S1 in supplementary material). In the treated groups by gemcitabine
hydrochloride or Gem-C12, all groups except Gemcitabine iv (i.e., Gem-C12 iv, Gem-C12 LNC
iv and Gem-C12 LNC sc) showed significant differences (p < 0.05, log rank test) in comparison
with the saline iv group (Table S1 in supplementary material). However, Gem-C12 LNC sc showed
a significantly prolonged lifespan (p < 0.05, log rank test) in comparison to its control group (i.e.,
non-loaded LNC sc); and Gem-C12 LNC iv showed a significantly prolonged lifespan (p < 0.05,
log rank test) in comparison to the Gemcitabine iv group (Table S1 in supplementary material).
The weight evolution of the different mice groups remained similar during the experiment with a
short stabilization during the treatment period (Figure 2-B).
Visualization of fluorescent LNC-based formulations in lung and mediastinal lymph
nodes
The in vivo behavior of LNC-based formulations was then assessed following the iv injection of
DiD-LNC and the sc injection of DiD-Gem-C12 LNC in tumor-bearing SCID-CB17 mice to
better understand the result of antitumor efficacy. The fluorescence images obtained 4h after
DiD-loaded LNC formulation injections are presented in Figure 3. DiD LNC is visible in the
entire lung of the three mice after iv injection (Figure 3-B) because at this time, the LNC mainly
remained in the blood as confirmed by plasmatic exposition (Figure 1-B) and because the lung is
a highly vascularized organ. However, no accumulation of DiD LNC (after iv administration) was
seen after 4h in mediastinal lymph nodes contrary to the other lymph nodes during the
biodistribution study (Figure 1-E). For DiD-Gem-C12 LNC administered by sc route, only an
intense local accumulation in mediastinal lymph nodes was observed (Figure 3-D) and not in the
entire lung of the three mice. Mediastinal lymph nodes are known to be close to the thymus [30].
16
Indeed, a negligible plasma exposition for DiD-Gem-C12 LNC was previously seen in Figure 1B.
Tolerance of the LNC-based gel technology on SCID-CB17 mice
To evaluate if the different treatments used in SCID-CB17 mice induced myelosuppresion, the
complete blood count was performed 8 days after the first injection as in clinic [31] and are
presented in Table 2. A significant decrease of platelet count was only observed with the group
treated by gemcitabine hydrochloride iv in comparison to the saline iv control group (p < 0.05,
Kruskal-Wallis test) and no difference was measured for Gem-C12 non or loaded in LNC and
delivered iv or sc in the respective schedule. SCID-CB17 mice are severe combined
immunodeficiency mice characterized by severe lymphopenia [32], which prevents the analysis
of the compete granulocyte count. Plasma biochemical parameters such as aspartate
aminotransferase (ASAT), alanine aminotransferase (ALAT), alkaline phosphatase (PH ALK)
and serum creatinine were evaluated (Table 3). No significant difference was seen between the
different groups in comparison to the control saline iv group, except for a significant decrease of
PH ALK for Gem-C12 not loaded in LNC, but formulated with ethanol as a co-solvent and in
surfactant micelles (p < 0.05, Kruskal-Wallis test).
Discussion
Drug delivery to the lymphatic system is a good option to combat metastasis spreading from a
primary tumor via the lymphatic system and to modulate immunity. Indeed, the lymphatic system
filters particles (e.g., nanomedicine) or cells (e.g., cancer cells) from interstitial fluid and supports
the activity of the lymphocytes, which furnish immunity, or resistance, to the specific disease17
causing agents [18]. In this study, using the sc administration route, slightly negative-charged
50nm LNCs (polyethene glycol 660-based surface) composing the hydrogel were chosen because
they could be the most appropriate method for lymphatic drainage and lymph node retention [5,
14-19]. The gel form of DiD-Gem-C12 LNC progressively released 50nm LNCs in the interstitial
fluid. Released LNCs passed rapidly into the lymphatic capillary vessels and were drained until
lymph nodes as observed by the specific accumulation in adjacent lymph nodes: axillary,
brachial, cervical (Figure 1-F) and mediastinum lymph nodes (Figure 3-B). When DiD-LNCs
were delivered intravenously, the LNCs were taken up by the mononuclear phagocytic system
that recognized the adsorption of complement proteins on the LNC surface, as observed by their
rapid accumulation in the liver and spleen (Figure 1-C), and as previously reported by Hirsjarvi et
al [25, 33]. DiD-LNC was also partly carried up to lymph nodes, as revealed by their
accumulation in all the studied lymph nodes (Figure 1-E). However, they were not accumulated
in mediastinum lymph nodes, where the interstitial fluid of the lung is drained. Indeed, healthy
blood vessels of the lung have poor permeability due to small pores in postcapillary venules (<
6nm) [34]. As a consequence, nanomedicine such as 50nm-LNC was poorly extravasated in lung
parenchyma and then poorly transported to mediastinum lymph nodes. LNC-based gel
technology, able to target the lymph nodes more specifically by the subcutaneous route and
loaded with a pro-drug of gemcitabine, Gem-C12, was tested on a patient-like lymphogenous
metastatic Ma44-3 preclinical model. There are very few preclinical models of lymphogenous
metastatic NSCLC in the literature. The Ma44-3 preclinical model developed by Ishikura et al
[20] was chosen because i) it is a subpopulation of a human squamous NSCLC grafted in the lung
[20], ii) it spreads in the mediastinum [20, 35], and iii) it causes the death of mice from
mediastinum metastases and not from the primary tumor [36].
18
Despite the presence of the primary lung tumor responsible for cancer cells spreading to the
mediastinum, Gem-C12 LNC only localized in mediastinum lymph nodes were able to exert the
same antitumor efficacy as the systemic administration of Gem-C12 (loaded in LNCs or micelles)
(Figure 2-A), which exerted a systemic anticancer activity (i.e., on primary lung tumor and
therefore on the spreading of mediastinal metastases) and resulting in the increase of the survival
lifespan to the same extent (Figure 2A). Moreover, this similar survival improvement by targeting
only lymph nodes was also accompanied by lower systemic side effects. Indeed, Gem-C12 LNC
delivered by iv or sc routes presented no significant depletion in platelet count, a marker of
myelosuppression, contrary to conventional gemcitabine hydrochloride iv (Table 2), and no
significant depletion in alkaline phosphatase, contrary to Gem-C12 loaded in surfactant-based
micelles (Table 3) when compared to the control saline iv group, respectively. Consequently, the
developed LNC-based gel technology is able to delay death when delivered subcutaneously by
locally targeting the adjacent lymph nodes without leading to myelosuppresion as encountered
with conventional chemotherapy. Nowadays, market-approved nanomedicine offers a better
therapeutic index, mainly by drastically decreasing toxicity. For example, Doxil® (for the US)
and Caelyx® (for Europe), (pegylated liposomal doxorubicin), and Myocet®, a non-pegylated
liposomal doxorubicin, drastically decrease cumulative dose-dependent cardiotoxicity, but do not
improve survival in comparison to conventional doxorubicin [37]. The nab-paclitaxel
(Abraxane®), albumin-bound paclitaxel, overcomes toxicity and complications encountered with
the conventional formulation based on polyethoxylated castor oil (Cremophor® EL) and ethanol,
that allows an increase of the administrable doses to slightly improve survival [38].
Concerning immunomodulation, the preclinical model used in this study, despite the development
of metastases in the mediastinum which is similar to what happens in NSCLC patients, was based
19
on severe immunodeficiency. Indeed, SCID-CB17 mice present mutations at the kinase protein,
DNA-activated, catalytic polypeptide (Prkdcscid) locus responsible for non-mature T and B cells
[39]; these are necessary to assure successful human tumor cell grafting. Therefore, the
immunologic action of gemcitabine was not taken into account in the in vivo antitumor activity
study. Indeed, gemcitabine was shown to increase the expression of the major histocompatibility
complex (MHC) on malignant cells, enhancing the cross-presentation of tumor antigens to CD8+
T cells, and selectively killing myeloid-derived suppressor cells, both in vitro and in vivo, thus
facilitating T cell-dependent anti-cancer immunity [40]. Good results have been obtained
combining gemcitabine and immunotherapy in murine model of pancreatic carcinoma and in
non-resectable pancreatic patients during Phase II clinical trials [41, 42]. Therefore further
investigations are required to evaluate the impact of this localized treatment in lymph nodes to
enhance immunity against tumors and metastases.
Conclusions
Gel technology based on lipid nanocapsules loaded with a lipophilic pro-drug of gemcitabine
injected subcutaneously was able to target the adjacent lymph nodes, such as the mediastinal
lymph nodes, and to combat locally the metastases by displaying a survival increase to an
equivalent extent to that of a treatment delivered intravenously which stops the primary tumor
from spreading the metastases (i.e., Gem-C12 loaded in LNCs or micelles, both delivered
intravenously). Moreover, LNCs loaded with Gem-C12 and delivered iv or sc did not induce
myelosuppression, unlike the conventional systemic gemcitabine hydrochloride.
20
References
1.
Siegel R, Ma J, Zou Z and Jemal A. Cancer statistics, 2014. CA Cancer J Clin 2014; 64:
9-29.
2.
IARC. 2008 [updated 11/08/2014]; Available from: http://globocan.iarc.fr.
3.
Molina JR, Yang P, Cassivi SD, Schild SE and Adjei AA. Non-small cell lung cancer:
epidemiology, risk factors, treatment, and survivorship. Mayo Clin Proc 2008; 83: 584-94.
4.
Howlader N, Noone AM, Krapcho M, Neyman N, Aminou R, Waldron W et al. SEER
Cancer Statistic Review, 1975-2008, National Cancer Institute. 2010; Available from:
http://seer.cancer.gov/archive/csr/1975_2008/.
5.
Ryan GM, Kaminskas LM and Porter CJ. Nano-chemotherapeutics: Maximising
lymphatic drug exposure to improve the treatment of lymph-metastatic cancers. J Control
Release 2014; 193: 241-56.
6.
Detterbeck FC, Boffa DJ and Tanoue LT. The new lung cancer staging system. Chest
2009; 136: 260-71.
7.
Fung SF, Warren GW and Singh AK. Hope for progress after 40 years of futility? Novel
approaches in the treatment of advanced stage III and IV non-small-cell-lung cancer: Stereotactic
body radiation therapy, mediastinal lymphadenectomy, and novel systemic therapy. J Carcinog
2012; 11: 20.
8.
Gompelmann D, Eberhardt R and Herth FJ. Advanced malignant lung disease: what the
specialist can offer. Respiration 2011; 82: 111-23.
9.
Lepper PM, Ott SR, Hoppe H, Schumann C, Stammberger U, Bugalho A et al. Superior
vena cava syndrome in thoracic malignancies. Respir Care 2011; 56: 653-66.
10.
Ryan GM, Kaminskas LM, Bulitta JB, McIntosh MP, Owen DJ and Porter CJ. PEGylated
polylysine dendrimers increase lymphatic exposure to doxorubicin when compared to PEGylated
liposomal and solution formulations of doxorubicin. J Control Release 2013; 172: 128-36.
11.
Moysan E, Bastiat G and Benoit JP. Gemcitabine versus Modified Gemcitabine: a review
of several promising chemical modifications. Mol Pharm 2013; 10: 430-44.
12.
Toschi L and Cappuzzo F. Gemcitabine for the treatment of advanced nonsmall cell lung
cancer. Onco Targets Ther 2009; 2: 209-17.
13.
Moysan E, Gonzalez-Fernandez Y, Lautram N, Bejaud J, Bastiat G and Benoit JP. An
innovative hydrogel of gemcitabine-loaded lipid nanocapsules: when the drug is a key player of
the nanomedicine structure. Soft Matter 2014; 10: 1767-77.
14.
Trubetskoy VS, Cannillo JA, Milshtein A, Wolf GL and Torchilin VP. Controlled
delivery of Gd-containing liposomes to lymph nodes: surface modification may enhance MRI
contrast properties. Magn Reson Imaging 1995; 13: 31-7.
15.
Xie Y, Bagby TR, Cohen MS and Forrest ML. Drug delivery to the lymphatic system:
importance in future cancer diagnosis and therapies. Expert Opin Drug Deliv 2009; 6: 785-92.
16.
Chen J, Yao Q, Li D, Zhang B, Li L and Wang L. Chemotherapy targeting regional
lymphatic tissues to treat rabbits bearing VX2 tumor in the mammary glands. Cancer Biol Ther
2008; 7: 721-5.
17.
Oussoren C and Storm G. Liposomes to target the lymphatics by subcutaneous
administration. Adv Drug Deliv Rev 2001; 50: 143-56.
18.
Nishioka Y and Yoshino H. Lymphatic targeting with nanoparticulate system. Adv Drug
Deliv Rev 2001; 47: 55-64.
21
19.
Kaminskas LM and Porter CJ. Targeting the lymphatics using dendritic polymers
(dendrimers). Adv Drug Deliv Rev 2011; 63: 890-900.
20.
Ishikura H, Kondo K, Miyoshi T, Kinoshita H, Hirose T and Monden Y. Artificial
lymphogenous metastatic model using orthotopic implantation of human lung cancer. Ann
Thorac Surg 2000; 69: 1691-5.
21.
Heurtault B, Saulnier P, Pech B, Proust JE and Benoit JP. A novel phase inversion-based
process for the preparation of lipid nanocarriers. Pharm Res 2002; 19: 875-80.
22.
Calvo P, Gouritin B, Chacun H, Desmaele D, D'Angelo J, Noel JP et al. Long-circulating
PEGylated polycyanoacrylate nanoparticles as new drug carrier for brain delivery. Pharm Res
2001; 18: 1157-66.
23.
Allard E, Passirani C, Garcion E, Pigeon P, Vessieres A, Jaouen G et al. Lipid
nanocapsules loaded with an organometallic tamoxifen derivative as a novel drug-carrier system
for experimental malignant gliomas. J Control Release 2008; 130: 146-53.
24.
Peltier S, Oger JM, Lagarce F, Couet W and Benoit JP. Enhanced oral paclitaxel
bioavailability after administration of paclitaxel-loaded lipid nanocapsules. Pharm Res 2006; 23:
1243-50.
25.
Hirsjarvi S, Dufort S, Gravier J, Texier I, Yan Q, Bibette J et al. Influence of size, surface
coating and fine chemical composition on the in vitro reactivity and in vivo biodistribution of
lipid nanocapsules versus lipid nanoemulsions in cancer models. Nanomedicine 2013; 9: 375-87.
26.
Hirsjarvi S, Dufort S, Bastiat G, Saulnier P, Passirani C, Coll JL et al. Surface
modification of lipid nanocapsules with polysaccharides: from physicochemical characteristics to
in vivo aspects. Acta Biomater 2013; 9: 6686-93.
27.
David S, Carmoy N, Resnier P, Denis C, Misery L, Pitard B et al. In vivo imaging of
DNA lipid nanocapsules after systemic administration in a melanoma mouse model. Int J Pharm
2012; 423: 108-15.
28.
Texier I, Goutayer M, Da Silva A, Guyon L, Djaker N, Josserand V et al. Cyanine-loaded
lipid nanoparticles for improved in vivo fluorescence imaging. J Biomed Opt 2009; 14: 054005.
29.
Bastiat G, Pritz CO, Roider C, Fouchet F, Lignieres E, Jesacher A et al. A new tool to
ensure the fluorescent dye labeling stability of nanocarriers: a real challenge for fluorescence
imaging. J Control Release 2013; 170: 334-42.
30.
Van den Broeck W, Derore A and Simoens P. Anatomy and nomenclature of murine
lymph nodes: Descriptive study and nomenclatory standardization in BALB/cAnNCrl mice. J
Immunol Methods 2006; 312: 12-9.
31.
Gemcitabine. Gemcitabine FDA-approved label. 2005 [updated 10/02/2014]; Available
from: http://www.accessdata.fda.gov/drugsatfda_docs/label/2005/020509s033lbl.pdf.
32.
Custer RP, Bosma GC and Bosma MJ. Severe combined immunodeficiency (SCID) in the
mouse. Pathology, reconstitution, neoplasms. Am J Pathol 1985; 120: 464-77.
33.
Hirsjarvi S, Sancey L, Dufort S, Belloche C, Vanpouille-Box C, Garcion E et al. Effect of
particle size on the biodistribution of lipid nanocapsules: comparison between nuclear and
fluorescence imaging and counting. Int J Pharm 2013; 453: 594-600.
34.
Wang J, Lu Z, Gao Y, Wientjes MG and Au JL. Improving delivery and efficacy of
nanomedicines in solid tumors: role of tumor priming. Nanomedicine (Lond) 2011; 6: 1605-20.
35.
Ishikura H, Kondo K, Miyoshi T, Takahashi Y, Fujino H and Monden Y. Green
fluorescent protein expression and visualization of mediastinal lymph node metastasis of human
lung cancer cell line using orthotopic implantation. Anticancer Res 2004; 24: 719-23.
22
36.
Fujino H, Kondo K, Ishikura H, Maki H, Kinoshita H, Miyoshi T et al. Matrix
metalloproteinase inhibitor MMI-166 inhibits lymphogenous metastasis in an orthotopically
implanted model of lung cancer. Mol Cancer Ther 2005; 4: 1409-16.
37.
Leonard RC, Williams S, Tulpule A, Levine AM and Oliveros S. Improving the
therapeutic index of anthracycline chemotherapy: focus on liposomal doxorubicin (Myocet).
Breast 2009; 18: 218-24.
38.
Hawkins MJ, Soon-Shiong P and Desai N. Protein nanoparticles as drug carriers in
clinical medicine. Adv Drug Deliv Rev 2008; 60: 876-85.
39.
Shultz LD, Ishikawa F and Greiner DL. Humanized mice in translational biomedical
research. Nat Rev Immunol 2007; 7: 118-30.
40.
Galluzzi L, Senovilla L, Zitvogel L and Kroemer G. The secret ally: immunostimulation
by anticancer drugs. Nat Rev Drug Discov 2012; 11: 215-33.
41.
Ghansah T, Vohra N, Kinney K, Weber A, Kodumudi K, Springett G et al. Dendritic cell
immunotherapy combined with gemcitabine chemotherapy enhances survival in a murine model
of pancreatic carcinoma. Cancer Immunol Immunother 2013; 62: 1083-91.
42.
Yanagimoto H, Shiomi H, Satoi S, Mine T, Toyokawa H, Yamamoto T et al. A phase II
study of personalized peptide vaccination combined with gemcitabine for non-resectable
pancreatic cancer patients. Oncol Rep 2010; 24: 795-801.
23
Figure Captions
Figure 1. A: i) Extracted organs 72h after sc injection of DiD-Gem-C12 LNC (hydrogel); ii)
Fluorescence signal in the organs (10ms exposure at 630-800nm); iii) Scheme of extracted
organs; iv) Regions of interest defined for the extracted organs in order to semi-quantify the
amount of photons/cm2/s. B: DiD plasmatic concentration (µg/mL) vs time in healthy nude mice,
after () iv injection of DiD LNC in suspension and () sc injection of DiD-Gem-C12 LNC
hydrogel. C to F: Biodistribution of DiD LNC in healthy nude mice, after iv injection of DiD
LNC in suspension (C and E) and sc injection of DiD-Gem-C12 LNC hydrogel (5% GemC12/Labrafac w/w) (D and F). Semi-quantification of the fluorescence signals of the removed
liver, spleen and inguinal, axillary, cervical and brachial lymph nodes (LN) using fluorescence
images (10ms integration time) at various times post LNC administration (1, 4, 8, 48, 96 and
336h). (n=5, mean  SD).
Figure 2. A: Kaplan-Meier survival and B: weight evaluation curves of 10 SCID-CB17 mice
grafted with 2×104 Ma44-3 cells (implanted in the left lobe) and randomized on Day 0. The
negative controls were iv saline treated group (), iv non-loaded LNC treated group () and sc
non-loaded LNC treated group (). The groups were treated by gemcitabine hydrochloride ()
or Gem-C12 loaded in micelles () or loaded in LNC (), delivered by iv route; or Gem-C12
loaded in LNC (), delivered by sc route. The dose was 40mg (molar equivalent gemcitabine
hydrochloride) per kilo of body weight beginning on Day 5 post-tumor graft. The treatments were
administered iv (continuous line) via the tail vein three times a week or by sc (dashed line) twice
a week.
24
Figure 3. Visualization of the lungs and mediastina of three SCID-CB17 mice on Day 5 after
tumor implantation, 4 hours after oral administration of 400mg/kg of 5-aminolevulinic acid and
either 4h after iv administration of DiD LNC using A: Pp IX or B: DiD visualization mode; or sc
administration of DiD-Gem-C12 LNC using C: Pp IX or D: DiD visualization mode using a CRI
Maestro system. Pp IX mode revealed the lung (green-brown color), the lung tumor (slight red
color), the mediastinum (green color) and the thymus (red color). The DiD mode revealed LNC
distribution in the lung and the mediastinum. Arrows and circles in the picture indicate the
localization of detectable, mediastinal lymph nodes and tumor mass, respectively.
25
Tables
Table 1. Physicochemical characteristics of non-loaded LNC for iv and sc injection, Gem-C12
LNC for iv and sc injection, DiD LNC for iv injection and DiD-Gem-C12 LNC for sc injection.
(n=3; mean ± SD).
Gem-C12
ξ-Potc
Z-Avea
PdIb
(nm)
payload
Form
(mV)
(mg/mL)
a
Non-loaded LNC (iv)
68 ± 3
0.08 ± 0.02
-8 ± 3
---
Liquid
Non-loaded LNC (sc)
67 ± 2
0.07 ± 0.01
-10 ± 5
---
Liquid
Gem-C12 LNC (sc)
53± 2
0.07 ± 0.01
-7 ± 4
11 ± 0.2
Hydrogel
Gem-C12 LNC (iv)
53 ± 1
0.04 ± 0.01
-7 ± 3
2.75 ± 0.05
Liquid
DiD-Gem-C12 LNC (sc)
56 ± 2
0.04 ± 0.02
-5 ± 2
11 ± 0.2
Hydrogel
DiD LNC (iv)
55 ± 2
0.06 ± 0.02
-7 ± 4
---
Liquid
Z-Average, b Polydispersity index, c Zeta potential.
26
Table 2. Complete blood counts in different groups (n=5; mean ± SD, *: p < 0.05): WBC:
complete granulocyte count. RBC: red blood cell count. HGB: hemoglobin rate. HCT:
hematocrit. PLT: platelet count.
WBC
RBC
HGB
HCT
PLT
(giga/L)
(tera/L)
(g/dL)
(%)
(giga/L)
Saline iv
0.2 ± 0.1
9.8 ± 0.3
15.1 ± 0.3
47 ± 1
469 ± 50
Gemcitabine iv
0.2 ± 0.1
9.2 ± 0.3
14.0 ± 0.4
45 ± 1
245 ± 63 *
Gem-C12 iv
0.3 ± 0.2
9.1 ± 0.6
14 ± 1
45 ± 2
275 ± 63
Non-loaded LNC iv
0.6 ± 0.3
9.8 ± 0.2
15.1 ± 0.3
47.7 ± 0.9
465 ± 44
Gem-C12 LNC iv
0.14 ± 0.03
8.9 ± 0.2
13.7 ± 0.3
43.1 ± 0.9
284 ± 24
Non-loaded LNC sc
0.4 ± 0.1
9.4 ± 0.7
14 ± 1
46 ± 3
453 ± 20
Gem-C12 LNC sc
0.26 ± 0.06
9.4 ± 0.5
14.5 ± 0.8
45 ± 2
281 ± 83
27
Table 3. Biological analysis in different groups (n=5; mean ± SD, *: p < 0.05): ASAT: aspartate
aminotransferase. ALAT: alanine aminotransferase. PH ALK: alkaline phosphatase.
ASAT
ALAT
PH ALK
Creatinine
(UI/L)
(UI/L)
(UI/L)
(mg/L)
Saline iv
138 ± 115
24 ± 8
146 ± 8
1.8 ± 0.4
Gemcitabine iv
98 ± 27
27 ± 12
129 ± 17
3±3
Gem-C12 iv
152 ± 95
23 ± 8
97 ± 33 *
2±1
Non-loaded LNC iv
77 ± 20
22 ± 4
139 ± 4
1.7 ± 0.5
Gem-C12 LNC iv
82 ± 21
18 ± 7
143 ± 14
3±2
Non-loaded LNC sc
88 ± 52
24 ± 16
109 ± 5
2±1
Gem-C12 LNC sc
157 ± 52
32 ± 10
121 ± 11
4±3
28
Figures
Figure 1.
29
Figure 2.
C
30
Figure 3.
31