MESENCHYMAL STEM CELL DERIVED EXTRACELLULAR VESICLES LOADED WITH AT LEAST ONE PHOTOSENSITIZER AND USES THEREOF FOR THE TREATMENT OF PERITONEAL CARCINOMATOSIS

Abstract

Several gastrointestinal and gynecological malignancies have the potential to disseminate and grow in the peritoneal cavity. The occurrence of peritoneal carcinomatosis (PC) has been shown to significantly decrease overall survival in patients. Treatment of residual microscopic disease remains a challenge with new anticancer modalities development. Now, the inventors propose an innovative therapeutic management of peritoneal carcinomatosis (PC) that is bio-inspired and tumor-targeted by engineering MSC-derived EVs to encapsulate a photosensitizer (mTHPC) for improved photodynamic therapy efficiency and safety. In this work, the inventors first evaluated the biodistribution of EVs-mTHPC in a murine PC model and highlighted superior accumulation of mTHPC in the tumor compared to other mTHPC formulations (free drug and liposomal one (Foslip®). The effectiveness of PDT mediated by mTHPC vectorized in EVs has then been evaluated in PC. In accordance with pharmacokinetics, the results revealed both an enhanced light-induced therapeutic efficiency in terms of tumoral cytotoxicity, safety for surrounding tissue after laser irradiation, immunomodulation and improved survival time. Thus, the present invention relates to mesenchymal stem cell derived extracellular vesicles loaded with at least one photosensitizer and uses thereof for the treatment of peritoneal carcinomatosis.

Claims

1. An isolated mesenchymal stem cell derived extracellular vesicle loaded with at least one photosensitizer.

2. The isolated mesenchymal stem cell derived extracellular vesicle of claim 1, wherein the at least one photosensitizer is selected from the group consisting of porphyrins, hydroporphyrins, chlorins, bacteriochlorins, purpurins, porphycenes, verdins, cyanines, merocyanines, phthalocyanines, chloroaluminum and phthalocyanines.

3. The isolated mesenchymal stem cell derived extracellular vesicle of claim 1 wherein the at least one photosensitizer is meta-tetra hydroxyphenylchlorin (mTHPC).

4. A population of mesenchymal stem cell derived extracellular vesicles (MSC-EVs) according to claim 1.

5. A method of preparing the population of claim 4 comprising the steps of i) causing a turbulent flow of a culture medium in a container, wherein the culture medium comprises mesenchymal stem cells adhering to the surface of microcarriers, the microcarriers being in suspension in the culture medium, and wherein the culture medium also comprises an amount of the at least one photosensitizer, and then ii) collecting the population of mesenchymal stem cell derived extracellular vesicles from the culture medium.

6. (canceled)

7. A method reducing tumor cell growth and/or proliferation in a subject in need thereof comprising the steps of i) administering to the subject a therapeutically effective amount of the population of MSC-EVs of claim 4; and ii) light-activating photosensitizer loaded on the extracellular vesicles to produce cytotoxic species, wherein the cytotoxic species inhibit the tumor cell growth and/or proliferation.

8. A method of treating cancer occurring in body cavity in a subject in need thereof, comprising the steps of i) administering to the body cavity of the subject a therapeutically effective amount of the population of MSC-EVs of claim 4; and ii) light-activating photosensitizer loaded on the MSC-EVs to produce cytotoxic species, wherein the cytotoxic species inhibit tumor cell growth and/or proliferation in the body cavity, thereby treating the cancer.

9. The method of claim 8, wherein the cancer occurring in body cavity is peritoneal carcinomatosis or pleural metastasis.

10. The method of claim 9 wherein the peritoneal carcinomatosis results from a colorectal cancer or an ovarian cancer.

11. The method of claim 9 wherein the peritoneal carcinomatosis is pseudomyxoma peritonei.

12. The method of claim 8 wherein step ii) is performed via coelioscopy, laparoscopy, or thoracoscopy.

13. The method of claim 8 wherein steps i) an ii) are repeated at least 2, 3, 4, or 5 times.

14. A pharmaceutical composition comprising an amount of the population of claim 4.

15. The isolated mesenchymal stem cell derived extracellular vesicle of claim 2, wherein the phthalocyanines are phthalocyanines with metal substituents or phthalocyanines without metal substituents.

Description

FIGURES

[0067] FIG. 1. Experimental set-up for mTHPC-EV production by turbulence.

[0068] A. Schematic picture of the setup of 3D culture cell and optical micrograph of microcarriers carrying MSC at confluence (A1) and one isolated imaged with a epiflurescence imaging microscope (A2-3-4, with respectively DAPI, mTHPC and the two channels). Size poydispersion was measured by NTA (B). Size distribution histograms obtained by nanoparticle tracking analysis for turbulence EVs. A single EV is illustrated with SEM (C), with a scale bar=100 nm.

[0069] FIG. 2. Biodistribution of mTHPC in colorectal and ovarian murine model of peritoneal carcinomatosis.

[0070] Tissue distribution of mTHPC after intraperitoneal injection at 4 h, 15 h, 24 h and 48 h depending on vectorization type. Results are presented as mTHPC mass concentration in each organ (ng mTHPC/mg tissue) (left) and mTHPC tumoral selectivity is highlighted by the ratio tumoral/tissue concentration (right) in colorectal carcinomatosis for (A) free mTHPC, (B) EVs-mTHPC, (C) Foslip® and (D) in ovarian carcinomatosis for EVs-mTHPC. (E) Direct comparison of mTHPC concentration at t=24 h for all the investigated organs with the 3 types of vectorization. (F) Direct comparison of mTHPC concentration in tumor, liver, kidneys and bowel at 4, 15, 24 and 48 h post-administration depending on mTHPC formulation: free mTHPC, Foslip®, EVs-mTHPC n=4 mice per time point and vectorization type; data are represented as mean±SEM (two-way ANOVA, * p=0.02, *** p<0.0001).

[0071] FIG. 3. Model of peritoneal carcinomatosis, experimental set-up and EVs-mTHPC dose determination.

[0072] Picture of (A) representative intraperitoneal disseminated nodules of CT26 in mouse abdomen and experimental set-up for PDT with laser beam focusing on the peritoneal cavity (B-C): mice underwent a laparotomy before illumination (λ=650 nm at 0.1 W/cm.sup.2 for 100 sec). The peritoneum and the skin were separately sutured immediately after irradiation. The number of mice (without or with carcinomatosis) dead in the following 24 h post treatment is summarized in the table (D). The dose of 0.15 mg/kg (half of the lethal dose) was identified as suitable: mice responded positively to the treatment and histological analysis of nodules displays necrosis (E2) contrary to 0.05 mg/kg which induces no sign of necrosis (E1) (HE staining at 20× magnification). Liver injury with hepatocytes necrosis (area below dotted line, F) was observed after PDT at 0.30 mg/kg EVs-mTHPC but not at 0.15 mg/kg.

[0073] FIG. 4: Outcome of PDT treatments in colorectal PM as function of the PS formulation. (A) Necrosis value of tumor nodules evaluated through H&E histological analysis: the necrosis value (0-4) reflects the spatial extent of necrosis (n=20-25 nodules per control group and n=20 tumor nodules per PDT group, Kruskal Wallis test, *** p<0.0001). (B) Apoptosis evaluated using a TUNEL assay (n=6 per control group and n=10 per PDT group, Mann-Whitney non-parametric test, *p=0.04, **p=0.006) (C). Representative fluorescence microscopy of tissue cryosection stained with anti-CD31 antibody (scale bar=200 μm) (A: no treatment, B: laser, C: mTHPC, D: EVs-mTHPC, E: mTHPC+laser, F: EVs-mTHPC+laser) and (G) Quantification of CD31 fluorescence signal (n=15 per treatment, *p=0.03, **p=0.003, *** p<0.0001). (D). Quantification of Ki67 tumor proliferation index and immune cell infiltration in tumors: F4/80+macrophages, CD8+(E) and CD3+. T cells (n=30/group for KI67, F4/80 and CD3 analysis; and n=45/group for CD8 analysis, Kruskal Wallis test, *p=0.03, **p=0.003, *** p<0.0001).

[0074] FIG. 5. Representative image to demonstrate the volumetric analysis method with [18F]FDG TEP scan and global carcinomatosis evaluation using PCI.

[0075] A. A spherical volume of interest was drawn to encompass the whole hypermetabolic lesion in the FDG PET image. PET/CT fusion images were used to identify anatomical structures and to exclude colic activity. Representative photographs of tumors after second [18F]FDG TEP scan imaging. Nodes on parietal and diaphragmatic peritoneum was detected with an important sensitivity. B. Quantification of maximal standardized uptake value of peritoneal tumor and correlation between tumoral SUVmax after treatment/SUVmax before treatment. The development of the peritoneal carcinomatosis was slowed by PDT, n=5 mice per group, ** p=0.008. C. Peritoneal Carcinomatosis Index was significantly lower in PDT group, n=5 mice per group, *p=0.01.

[0076] FIG. 6. Kaplan-Meier survival curves for tumor-bearing mice and statistical comparisons of survival times after PDT.

[0077] A. Treatment groups (n=10 per group) included animals that received vectorized (or not) intraperitoneal photodynamic treatment (mTHPC 0.15 mg/kg, laser 10 J/cm2). Control groups (n=11 per group) included animals that received no treatment, laser illumination (10 J/cm2) without photosensitizer, mTHPC or EVs-mTHPC without laser illumination. B. Comparison between three PDT groups and no treatment group. C. A log rank test was carried out on survival data.

[0078] FIG. 7: Representative histological analysis of liver and kidneys 48 hrs after PDT.

[0079] Arrow indicate pathological changes in cell morphology (destruction of cell nucleus). Scale bars represent 120 μm for livers and 60 μm for kidneys.

EXAMPLE

[0080] Methods

[0081] a/ Production of MSCm-EVs Encapsulating mTHPC Photosensitizer (EVs-mTHPC)

[0082] Cell Culture in Flasks

[0083] Murine mesenchymal stem cells (MSCm) were cultured in DMEM at 37° C. and 5% CO.sub.2 in DMEM supplemented with 10% fetal bovine serum and 100 U/mL penicillin-steptomycin.

[0084] 3D Cell Culture in Spinner Flask Bioreactors and EVs-mTHPC Production

[0085] MSCm were trypsinized, rinced with PBS, seeded in spinner flask bioreactor in DMEM complete medium containing 5 g of 200 μm Cytodex 1 dextran microcarriers (GE Healthcare); they were then submitted to 24 cycles of 45 min of rest interspersed with 3 minutes of gentle mixing at 30 RPM to ensure homogeneous adhesion of cells on microcarriers. After cell adhesion, continuous gentle mixing was performed until cell confluence on microcarriers (3-4 days). Meta-tetra(hydroxyphenyl)chlorin (mTHPC) (Biolitec, Germany) was added to the culture medium to obtain a final concentration of 100 μM for an overnight labelling. Before EV production was launched, complete medium+mTHPC was rinsed with 5 washing steps (with serum-free white DMEM medium, 100 U/mL penicillin-steptomycin) were performed to remove the non-encapsulated drug. Spinner flasks were then submitted to mixing at 144 RPM during 4 hours for EVs-mTHPC production.

[0086] EVs-mTHPC Isolation and Purification

[0087] After EV production, EVs-mTHPC were isolated from the conditioned culture medium. First, microcarriers and cell debris were eliminated by 2,000 g centrifugation for 10 min. The following ultracentrifugation at 100,000 g for 70 minutes allowed isolating the EVs-mTHPC, which were then resuspended in phosphate-buffered saline.

[0088] EVs-mTHPC Characterization (Size, Yield and mTHPC Quantification)

[0089] EV size distribution and concentration were determined by Nanoparticle Tracking Analysis (NTA) using a Nanosight $NS300 HS with a 405 nm laser. Before measurements, EVs were diluted to an appropriate concentration (between 3×10.sup.8 and 2×10.sup.9) with sterile PBS (confirmed to be particle-free by NTA measurement). For each sample, 5 movies of 30 s were recorded using a camera level of 16. Data were analyzed with NTA Analytical Software. The concentration of m-THPC in samples of purified EVs was determined by fluorescence spectroscopy. An EnSpire (Perkin Elmer) plate reader spectrometer was used at 410 nm excitation wavelength. The drug concentration was obtained from fluorescence emission at 655 nm based on a standard calibration curve of m-THPC. For m-THPC quantification in EVs, Triton X-100 was added at 0.3% final concentration in order to lyse EVs.

[0090] b/ Cell Culture

[0091] Mouse colon (CT26) and ovarian (ID8) cancer cell line genetically modified to stably express luciferase (CT26-Luc and ID-8) were used. The CT26-Luc cells were cultured in Dulbecco's Modified Eagle's Medium (DMEM) with L-Glutamine supplemented with 10% fetal bovine serum, 1% penicillin-streptomycin at 37° C. in 5% CO2 humidified atmosphere. The ID8-Luc cells were cultured in Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 4% fetal bovine serum, 1% penicillin-streptomycin, 1% d′ITS, 1% de L-glutamine and 2.5 μg/mL of puromycin.

[0092] c/ In Vitro Experiments: Exposition to EVs-mTHPC and Laser Irradiation

[0093] The CT26-Luc and ID8-Luc cells were seeded in 24-well plates and incubated overnight. After PBS rinsing, the cells were incubated with XX μL of EVs-mTHPC at XX or XX μM during 24 h in the dark. Wells were then washed with PBS, covered with white DMEM and irradiated individually using a 650 nm diode laser featuring a fiber delivery system. The optical fiber was fixed so that the laser spot covers precisely the surface of the well (1.9 cm.sup.2). Cells were exposed to at a light fluence of 10 J/cm.sup.2 (100 mW/cm.sup.2 for 100 s). Cells were incubated for 24 h before cytotoxic assessment by the Alamar Blue test (Invitrogen), according to the supplier's instructions.

[0094] d/ Scanning Electron Micrograph (SEM) Observation

[0095] SEM was first used to visualize extracellular vesicles (EVs) at CT26 and ID8 tumoral cell surface. Tumoral cells were cultured in 8 well-chambers removable (Ibidi, 80841) for one day. Cells were washed with PBS and we replace with complete medium only, or containing EVs-mTHPC. After 24 hours, two well-chambers were illuminated at 10 J/cm.sup.2. After 24 hours, cells were fixed 30 minutes with 4% paraformaldehyde and then further rinsed three times with PBS buffer. Dehydration was performed by rinsing the samples through graded ethanol/water mixtures (50%, 70%, 80%, 90%, and finally 100%, each step for 10 min at 4° C.).

[0096] e/ Animal Model

[0097] All animal experiments were performed in agreement with institutional animal use and care after approval by the local Ethics Committee (registration number for the carcinomatosis model experiment: APAFIS-8617). Six-week-old female BALB/c (provided by Charles River, Arbresle, France and weighing 18 g) and C57BL/6 mice were provided with food and water ad libidum, for colon and ovarian model respectively. The animals were allowed to acclimate to the facility for at least one week before being used for experiments. Colorectal model was induced with an intraperitoneal injection of 5.Math.10.sup.4 CT26 cells resuspended in a volume of 200 μL of physiological saline per inoculum. Ovarian model was induced with an intraperitoneal injection of 10.sup.6 ID8-LUC cells suspended in 1 mL of medium. Between the days of IP injection and sacrifice, the wellbeing of the mice was checked twice a week through the search of any sign of pain, dehydration, changing in behavior or loss of weight.

[0098] f/ In Vivo Photodynamic Therapy (PDT)

[0099] The mTHPC and EVs-mTHPC were injected intraperitoneally with 200 μL of free mTHPC (in a solution of ethanol/polyethylene glycol 400/water at a 2/3/5 volume ratio) or vectorized mTHPC with EVs or liposome (Foslip®), at 0.15 mg/kg drug concentration. mTHPC [3,30,300,3000-(2,3-dihydroporphyrin-5,10,15,20-tetrayl)-tetraphenol] and its liposomal formulation (Foslip®) were kindly provided by Biolitec research GmbH (Jena, Germany). After 24 h, animals from both groups were anesthetized with isoflurane, a midline laparotomy was performed to allow invasive laser irradiation, and were then irradiated using a 650 nm laser at a fluence of 10 J/cm.sup.2 (100 mW/cm.sup.2 for 100 s). Two irradiations spots were successively realized to allow an illumination of the entire cavity and drops of saline were used to keep the tissue moist. A heating device was used to keep body temperature stable throughout the treatment until the animal was woken up. The peritoneum and the skin were separately sutured immediately after irradiation.

[0100] g/ mTHPC Biodistribution: Drug Extraction and Spectroscopy Measurements

[0101] Drug extraction mTHPC levels in organs after IP injection of mTHPC, EVs-mTHPC or Foslip®, were quantified using chloroform extraction based on a previous protocol described in Foster et al. (Foster et al, Translational Oncology (2010) 3, 135-141). Briefly, tumor-bearing animals (with colorectal carcinomatosis) were injected at 0.5 mg/kg mTHPC intraperitoneally. Either at 4 h, 15 h, 24 h or 48 h mice were sacrificed to determine mTHPC accumulation in different organs: tumors, liver, spleen, kidney, intestine, peritoneum, skin and lung. First, organs were weighed and grinded in PBS (1 mL per 10 mg of organ). The amount of 0.8 mL of MeOH per 10 mg of organ was added to the solution and vortexed; then, 0.8 mL of chloroform was added. The solution was centrifuged at 1400 rpm for 10 minutes to allow the separation of the aqueous and chloroform phases. The bottom layer consisting of the chloroform and the solubilized mTHPC was transferred for spectroscopy measurements. Fluorescence emission spectra were obtained by a fluorescence spectrophotometer using 415 nm excitation wavelength. Biodistribution of EVs-mTHPC was compared in ovarian carcinomatosis.

[0102] Ascitis smears were performed as soon as the sacrifice at 24 h, in ovarian model. Ten minutes after 4% paraformaldehyde fixation, cells were permeabilized with 0.1% Triton for 5 minutes and then blocked with 1% BSA for one hour. The cells were then incubated overnight at +4° C. with anti-firefly Luciferase antibody [EPR17789] (AlexaFluor 488) (ref ab237251) at 1/100 dilution. Tumoral cells presented in ascitis were detected with immunofluorescence analysis with a confocal microscope (Carl Zeiss Microscopy GmbH LSM 800). Colocalisation between tumoral cells expressing LUC and mTHPC fluorescence was observed (excitation 405 nm, emission 635-700 nm).

[0103] h/ Safety, Tolerability and Preliminary Anti-Tumor Activity

[0104] The maximum tolerated dose of EVs-mTHPC for intraperitoneal PDT was evaluated. Mice with (n=10) and without colorectal peritoneal metastasis (n=9) received three concentrations of mTHPC: 0.05 mg/kg, 0.15 mg/kg and 0.3 mg/kg. Drug injection was performed 8 days after IP inoculation of CT26 LUC cells. PDT was performed the day after IP injection of EVs-mTHPC; a midline xiphoid-pubic laparotomy was made to allows intraperitoneal laser illumination. Mice were sacrificed three days after PDT. Tumoral metastasis, liver and kidney were preserved in 4% paraformaldehyde for histological analysis.

[0105] i/ Effectiveness and Toxicity of PDT with EVs-mTHPC in Mouse Model of Colorectal Peritoneal Carcinomatosis

[0106] Effectiveness evaluation of PDT with EVs-mTHPC was performed with two experimentations, to assess early and long-term effectiveness. Treatment injection was made 12 days after IP inoculation of CT26, followed by invasive illumination 24 hours after. Six groups were performed for each experimentation: (i) no treatment, (ii) laser, (iii) free mTHPC, (iv) EVs-mTHPC, (v) free mTHPC+laser, (vi) EVs-mTHPC+laser.

[0107] Early effectiveness was evaluated 48 hours after illumination to compare: tumoral necrosis, apoptosis, macrophages invasion, lymphocyte recruitment, vascular damage and proliferative index result according to treatment. Long-term effectiveness consisted on a survival analysis function of the treatment.

[0108] Lesion Characterization by Pathology

[0109] Tumor were fixed in 4% paraformaldehyde, embedded in paraffin and sectioned at 4 μm to histological processing using hematoxilin/eosin (HE) and immunohistochemical stains. Depth of necrosis was blindly determined for each slide by a confirmed pathologist. Necrosis was evaluated using a semi-quantitative histology. Necrosis Value (NV) was determined according to the depth of the necrosis for each tumor nodule. The NV was correlated to the percentage of necrosis: 0: no necrosis, <25%=1; 25-50%=2; 50-75%=3; >75%=4. This technique was used to evaluate toxicity on live and kidney too.

[0110] For immunohistochemical detection of the chosen markers for tumor tissue, sections were deparaffined and subject to antigen retrieval methods validated for each of the primary antibodies. Sections were incubated overnight at 4° C. with primary antibodies (anti CD8 antibody (Cell signaling technology #98941), anti CD3 antibody (ab5690), anti F4/80 antibody (Cell signaling technology, #70076) and anti-Ki-67 antibody (ab16667), then incubated with secondary antibody and developed using the avidin-biotin complex method with 3,3′diaminobenzidine as chromogen. Histology and immunostaining preparations were performed on the Cochin HistIM Facility, Paris.

[0111] Antibody expression for each analysis was evaluated by the mean number of positive cells in 3 randomly chosen areas (×40 magnification) for 10-15 nodes by treatment. Quantitative analysis of immunostaining preparation was performed using the color segmentation ImageJ plugin developed by the Biomedical Imaging Group at the EPFL, Switzerland.

[0112] Immunofluorescent Analysis of Tumor

[0113] TUNEL assay: Cell apoptosis in tumour was detected by a FragEL DNA fragmentation (TUNEL) detection kit (Sigma, Roche, ref 11684795910). In brief the cryostat section were permeabilized with 0.1% Triton and 0.1% (citrate de sodium) in 10×PBS. The slides were then labeled with a TdT reaction mixture for 60 min and were mounted with a mounting solution containing 4′, 6-diamidino-2-phenylindole (DAPI). The apoptotic cells (green) and cell nucleus (blue) was examined using a fluorescence microscopy. The percentage of apoptotic cells was assessed in 2 randomly fields at 40× magnification. The apoptotic index was calculated with image J.

[0114] For image-based quantitative analysis of blood vessels, immunofluorescence detection of CD31 was made after one hour incubation with antiCD31 antibody (BD 553370,1/50) and 45 min incubation with second antibody (Alexa Fluor® 488, ThermoFisher, A11006, dilution 1/200)

[0115] Survival Analysis

[0116] Another 10 or 11 mice were selected from each group to evaluate the overall survival until 30 days. The end point was defined as mice death or the cachexia with the loss of 20% of the total weight that required the killing of the animal (as mentioned in the registration).

[0117] j/ Metabolic Activity Response of Peritoneal Metastases Measured by [18F]FDG PET/CT Scan after PDT with EVs-mTHPC

[0118] Ten days after CT26 IP injection, five mice were IP injected with EVs-mTHPC with IP illumination 24 h after. [18F]FDG PET/CT scan imaging were done before and 48 hours after PDT. Metabolic activity of various metastatic lesions was measured. Maximal standardized uptake value (SUVmax) was quantitatively used to determine 18F-FDG avidity. SUV was defined as the concentration of 18F-FDG divided by the injected dose, corrected for the weight of the mice and radioactive decay at scanning time [SUV=activity concentration/(injected dose/mice weight)]. The results were compared with five control mice, without treatment. We correlate this result with clinical extension of metastasis by calculating peritoneal carcinomatosis index (PCI) score.

[0119] k/ Statistics

[0120] All data were analyzed with the GraphPad Prism® version 7 software. Results were represented mean±SEM. comparison was performed using either a Mann-Whitney non-parametric test, or a Kruskal-Wallis test depending on the number of groups to compare. The survival curves were calculated by the Kaplan-Meier method, and the statistical significance of differences in the cumulative survival curves between the groups was evaluated by logrank test. The P values less than 0.05 were considered significant.

[0121] Results

[0122] Photosensitizer Synthesis: EVs-mTHPC

[0123] After reaching confluence on beads (as exemplified in FIG. 1.A), MSCs were incubated overnight with mTHPC. FIG. 1.A.2-4 displays MSCs on a bead after mTHPC labeling and evidences the cytoplasmic localization of the internalized mTHPC. With an initial number of cells of 32.Math.10.sup.6 seeded on beads and after 3 days of cell division, the total quantity of EVs produced after mixing at 144 RPM during 4 hours was measured by NTA at 1.4×10.sup.13 EVs. Size distribution of EVs was analyzed both by NTA and SEM (FIG. 1.B-C): classical shape and typical polydisperse EV size range was observed, with a mean size of 175.2 nm±14.5. Diluted EV sample at 1.2×1012 EV/mL featured a concentration of 100 μM of mTHPC. To ensure EV stability throughout the duration of the study, EVs were stored at −80° C. No changes of EV shape nor mTHPC concentration was noticed.

[0124] In Vivo Biodistribution Distribution of mTHPC Function of Formulation

[0125] The mTHPC concentration as a function of time after injection is shown in FIG. 2. For all organs except the tumors, the error bars indicate the standard errors arising due to inter-animal variations. Whereas for tumor tissue, the error bars also partly reflect intra-animal differences.

[0126] In colorectal model of carcinomatosis, for tumor tissue, the mTHPC levels at 4 h and 24 h were higher for EVs nanovectorization than free or liposomal vectorization of mTHPC (4 h: 0.36±0.1 vs 0.10±0.01 or 0.15±0.04, and 24 h: 0.50±0.1 vs 0.16±0.05 or 0.06±0.01 ng/mg tissue respectively). At 15 h intra tumoral concentration were almost equivalent in three groups (0.19±0.02 vs 0.15±0.04 or 0.21±0.07 respectively) and at 48 h, we noted an elimination of mTHPC (0.10±0.04 vs 0.10±0.00 or 0.14±0.03 respectively). We observed for free mTHPC, the tumor-to-organ ratio did not change significantly with time and displayed a total average of 2.5. The average selectivity of mTHPC in tumor compared to other organs investigated is listed for the time points investigated, function of the mTHPC formulation. At 24 h, the tumor-to-organ ratio averaged was between 1 and 2 for free mTHPC and between 1 and 5 for liposomal formulation. The highest selectivity achieved with EVs formulation, with a ratio between 12 and 153 at 24 h. The biological nanovectorization of mTHPC allowed a 40 and 51 times higher selectivity in tumor-tissue compared to liposomal vectorization and free formulation respectively.

[0127] We analysed the EVs nanovectorization in a second model of carcinomatosis, from ovarian origin. The highest tumoral selectivity achieved at 24 h too, with a mean tumor-to-organ ratio at 15.

[0128] Safety, Tolerability and Preliminary Anti-Tumor Activity

[0129] The toxicity data for non-tumor bearing and tumoral mice treated with increasing doses of EVs-mTHPC is shown in FIG. 3. The majority of mice receiving an injection of 0.3 mg/kg EVs-mTHPC died less than 24 h after illumination: 75% (3 out of 4) of tumoral mice and 66% (2 out of 3) of non-tumoral mice. All mice injected with 0.05 and 0.15 mg/kg of mTHPC survived. Histological analysis of the liver indicated some histopathological changes after 0.30 mg/kg IP injection of EVs-mTHPC, with nuclear degradation and infiltration by inflammatory cells. The liver toxicity could explain death of mice. Tumoral histological analysis showed necrosis at 0.15 mg/kg, but no necrosis was detected at 0.05 mg/kg. 0.15 mg/kg appeared like the ½ lethal dose, with a preliminary antitumoral effectiveness.

[0130] Effectiveness of PDT in a Mice Model of Peritoneal Carcinomatosis from Colorectal Origin

[0131] A total of 94 tumors from control mice (n=27 untreated, n=23 laser, n=25 mTHPC, n=19 EVs-mTHPC) and 40 tumors from mice treated with PDT (n=20 per group) were analyzed by H&E staining (FIG. 3). PDT performed on carcinomatosis lesions at 650 nm with an irradiance 0.1 W/cm.sup.2 for 100 sec induced necrosis in 72.5% of cases versus 13% with control treatment (mean NV control=0±0.06, mean NV mTHPC+laser=1.6±0.3, mean NV EVs-mTHPC+laser=1±0.2), p<0001. Intra tumoral necrosis was not extensive but involved both the center and periphery of the node. To further investigate the mode of death in the observed tissue damage areas, we used TUNEL staining to assay for apoptosis revealing that tumors after EVs-mTHPC+laser had a higher level of green fluorescence than others treatment. The mean fluorescence per mm.sup.2 of tumor was significantly (p<0.05) higher after treatment with EVs-mTHPC+laser than in the other treatment groups: no treatment: 0.55±0.4, Laser: 0.54±0.8, mTHPC: 0.86±0.4, EVs-mTHPC: 1.42±0.4, mTHPC+laser: 2.17±0.2 versus EVs-mTHPC+laser: 11.60±2.2. We then compared the short-term outcome of PDT treatment on the colic PM 48 h after the single laser exposure for the three PS formulations. None of the formulations without laser irradiation induced tumor necrosis (dark toxicity) (FIG. 4A). Significant increases in necrosis values (NV) were observed in irradiated mice injected with the free drug (mTHPC+laser, mean NV=1.6±0.3) and with EVs-mTHPC (EVs-mTHPC+laser, NV=1±0.2) compared to control non exposed groups. Intra-tumor necrosis did not extend to the whole tumor volume but involved both the center and the periphery of the node (data not shown). Apoptosis level assessed by TUNEL was 5-fold higher in the laser-exposed group receiving EVs-mTHPC in comparison to the Foslip® and free drug (FIGS. 4B and 4D) and 36-fold higher than in control non-exposed groups. Proliferation index was reduced by a factor of 5 with photoactivated EVs-mTHPC versus 2 for the free drug and 3 for the Foslip® formulation.

[0132] To investigate the indirect anti-tumor effects of PDT on intraperitoneal dissemination, additional immunohistochemistry explored macrophages infiltration and lymphocyte recruitment. Macrophage intra-tumor invasion (anti F480 antibody) was significantly higher in mice treated with PDT (mTHPC+laser 41.45%±2.2, EVs-mTHPC+laser 41.04%±2.1) than control mice (untreated 19.34%±2.6; 25.78%±1.5, mTHPC 13.19%±1.4, EVs-mTHPC 23.26%±2.5), p<0.0001. Infiltrated inflammatory cells were seen in the necrosis zone among the PDT groups. The PM microenvironment was differently modified by the different treatments, particularly the immune cell tumor infiltrate. Noteworthy the tumor invasion of F4/80 macrophages (FIG. 4D) was significantly higher in mice treated with free drug or EVs-mTHPC and irradiated in comparison to the non-irradiated or laser only groups, but not in mice treated with Foslip® and irradiated. This result is in line with necrosis values that were maximal for PDT with free drug and EVs-mTHPC and could elicit tumor inflammation

[0133] The T-lymphocyte infiltration, unlike macrophage infiltration, was increased first by the EVs, and second by the PDT. Anti CD3 labeling in PDT treated mice after IP injection of EVs-mTHPC (4.14%±0.5) was twice as high than EVs-mTHPC (2.55%±0.3) and PDT group after free mTHPC IP injection (2.51%±0.3), in contrast with other control groups (untreated groups 0.71%±0.1; laser: 0.79%±0.1, mTHPC: 0.33%±0.07), p<0.0001. To explore defining T-cell category, we performed CD8 immunodetection, with the same results. CD8 detection in PDT treated mice after IP injection of EVs-mTHPC (4.40%±0.3) was twice as high than EVs-mTHPC (1.96%±0.2) and the PDT group after free mTHPC IP injection (1.88%±0.2), in contrast with other control groups (untreated groups 1.45%±0.13; laser 0.65%±0.08; mTHPC 0.44%±0.07), p<0.0001. Indeed, the number of CD3+ T cell markedly increased in tumors when both Foslip® or EVs-mTHPC were injected in comparison with the free drug and controls (FIG. 4D). However, laser irradiation amplified T-cell infiltration with the most prominent effect with EVs-mTHPC (FIG. 4D). The accumulation of effector CD8+ cell in tumors show a similar trend with increased recruitment in mice treated with EVs-mTHPC+laser (FIG. 4D). As the percentage of CD3+ and CD8+ within the nodules were equivalent, lymphocyte infiltration can be mainly composed of cytotoxic lymphocytes. Overall, PDT in colon PM induces a pro-inflammatory immune environment with inflammatory macrophages and cytotoxic T cell infiltration that is mostly promoted by mTHPC vectorization with MSC-derived EVs. Finally, a reduced expression of CD31 endothelial cell marker (FIG. 4C) after PDT, indicating vascular damages, with the most prominent effect in mice treated with EVs-mTHPC+laser. The percentage of CD3 and CD8 positive within the nodules being equivalent, lymphocyte infiltration was mainly composed of cytotoxic lymphocytes. Moreover, the lymphocyte infiltration increasing with EVs, probably mesenchymal nature of these vesicles permitted an immunomodulatory character.

[0134] Immunofluorescence staining further demonstrated that PDT treated tumors exhibited smaller CD31 (an endothelial cell marker) expression (mTHPC+laser: 0.55.Math.10.sup.7±1.0.Math.10.sup.6 and EVs-mTHPC+laser: 0.29.Math.10.sup.7±0.6.Math.10.sup.6) than control groups (no treatment 1.02.Math.10.sup.7±1.0.Math.10.sup.6, Laser: 1.09.Math.10.sup.7±1.4.Math.10.sup.6, mTHPC: 1.06.Math.10.sup.7±1.4.Math.10.sup.6, EVs-mTHPC: 0.92.Math.10.sup.7±1.9.Math.10.sup.6).

[0135] In total, PDT with mTHPC permitted an antitumoral action by necrosis, macrophages infiltration and vascular damage. mTHPC nanovectorization with EVs permitted a more cytotoxic PDT effect with apoptosis and lymphocyte recruitment.

[0136] In consequences, to further characterize the effect of PDT in peritoneal disseminated tumors proliferation, we performed immunohistochemical staining for Ki67. The mean Ki67-index was significantly lower in the PDT-treated group after injection of EVs-mTHPC (9.2%±1.4) compared to controls (no treatment 46.6%±2.2; Laser 52.7%±2.5; mTHPC 45.52±3.4; EVs-mTHPC 51.9%±2.3) and the group treated with PDT after free mTHPC injection (26.2%±4.5), p<0.0001.

[0137] Metabolic Activity Response of Peritoneal Metastases Measured by [18F]FDG PET/CT Scan after PDT with EVs-mTHPC

[0138] 9 mice received the 2 PET/CT scan images: 5 untreated and 4 treated. The injection of [18]FDG could not be performed correctly in one treated mouse, with a diffusion of the product at the level of the tail. The mean SUVmax ratio 2.sup.nd/1.sup.st PET/CT scan imaging was significantly higher in the control group compared to the treated group (1.94±0.1 vs 1.45±0.06, p=0.008 respectively). The SUVmax of the 2.sup.nd imaging was systematically higher than the 1.sup.st imaging, with apparition of tumors on the 2.sup.nd imaging. It shows the aggressiveness of this tumor model with exponential growth. PDT treatment did not block tumor development in our model but induced a slowdown. This treatment effectiveness was also expressed by the evaluation of PCI. Mean PCI was significantly higher in untreated mice compared to PDT-treated mice (16±0.8 vs 11±0.7 respectively, p=0.01). In addition, we showed that this imaging was very sensitive for infra-millimeter nodule in parietal peritoneal node (FIG. 5). However, PET/CT scan was not suitable for evaluating mesenteric tumor lesions in our model due to the intense hypermetabolism digestive tract despite the fasting period.

[0139] EVs-Nanovectorized PDT Leads to Significant Survival Advantage

[0140] At the end of the study, the survival rate of the EVs-mTHPC+laser group was 30%, with a median survival of 28 days, while there were no survivor in the other groups (median survival: 16, 20, 20.5, 22, 24, 24.5, and 26 days for mTHPC+laser, mTHPC, EVs-mTHPC, no treatment, laser, Foslip®, and Foslip®+laser respectively). The Kaplan-Meier survival curves of the different groups are shown in FIG. 6. PDT mediated by EVs was able to significantly prolong mice survival in comparison with others PDT-treated group and control groups. The P values of the log-rank test comparisons are shown in table. Non-vectorized mTHPC was the cause of lethal toxicity with 91% of dead mice 4 days after laser illumination. mTHPC vectorization with liposome (Foslip®) and EVs (EVs-mTHPC) allowed to suppress this lethal toxicity, and to prolong survival in comparison with control groups. Biological nanovectorization with EVs permitted a better survival than liposomal vectorization (p=0.005).

[0141] In Vivo Toxicology

[0142] In this study, the toxicity of mTHPC function of vectorization was systematically investigated in mice following intraperitoneal injection at 0.15 mg/kg. Mice were sacrificed at 72 h after treatment injection and 48 h after illumination. The body weight of the different groups of treatment was similar. Histological assessment was performed to examine tissue damage, mainly liver and kidney which are the most affected organs to iatrogenics. Representative histology results are shown in FIG. 7. In our study, accumulation of mTHPC in the liver after free mTHPC and Foslip®, caused adverse effects including pathological changes in their morphology with nuclear degradation. Histological analysis of the kidney (including the glomeruli) indicate some histopathological changes after IP injection of free mTHPC. These observations were explained with elemental analysis (FIG. 2) which showed relatively low amounts of mTHPC in organs after IP injection of EVs-mTHPC.

DISCUSSION

[0143] We report preclinical evidences to assess the high specificity of biological nanovectorization with EVs-targeted photosensitizer which could enable intraperitoneal photodynamic therapy for peritoneal carcinomatosis of colic and ovarian origin. In the 30 last years, with the development of new types of photosensitizers, PDT has attracted people's interest as a treatment method. The poor prognosis of peritoneal metastasis and recent developments in nanovectorization have generated considerable interest in PDT for this disease.

[0144] EVs appears like a natural drug delivery vehicles with negligible immunogenicity at contrary to synthetic nanovectors as liposomes [van Dommelen 2012]. Much better stability and intracellular accumulation was demonstrated for EVs-mTHPC compared to mTHPC liposomal formulation [Millard 18]. mTHPC embedding into EVs prevents PS aggregation, like liposomal nanovectors [Reshetov 2012], with a better tumoral vectorization and slower clearance. For tumor tissue, the mTHPC concentration with liposomal vectorization was the most important at 15 h (0.21±0.07 ng/mg tissue), following by a rapid clearance (at 24 h: 0.06±0.01 ng/mg tissue). Xie et al described too, with a liposomal formulation of mTHPC, a maximal mTHPC concentration of 0.17±0.07 ng/mg at 18 h post injection [Xie15]. At 4 h, we measured an intra tumoral concentration at 0.15±0.04 nm/mg tissue, which corresponds to the data of the literature (0.16±0.024 ng/mg tissue in Svensson analyze [Svensson06]. So we proposed a new generation of PS, with a biological vectorization, which allows a concentration of higher intra-tumor PS, and a more important tumor selectivity after IP injection. The tumor-to-normal tissue ratio allows understanding the high grade morbidity observed both in preclinical and clinical studies. Our analysis yielded a tumor-to-organ ratio of 19 at 4 h, 24 at 15 h, 44 at 24 h and 7 at 48 h, which was significantly higher than Foslip® (3 at 4 h, 5 at 15 h, 2 at 24 h and 6 at 48 h) and higher than literature ratio [Morlet95, Veenhuizen97]. The hydrophobic mTHPC (without vectorization) preferentially accumulated in organs rich in mononuclear phagocytic cells (e.g., liver and spleen) to a higher degree than in other types of tissues. In ovarian model of carcinomatosis, EVs permitted an tumoral vectorization too with a peak at 24 h. A mean tumor-to-normal tissue of 15, which is highest of recent PDT vectorisation in literature [Azaïs16], should permit to illuminate the peritoneal cavity without inducing visceral injuries. These results showed an important ratio regarding liver kidney and bowel (tumor-to-tissue ratio=12, 23 and 25 respectively), indicating that it could be possible to illuminate those organs with appropriate wavelength without risking an hepatic or kidney injury or digestive perforation. However these complications limited the development of PDT yet. We propose an IP injection of PS. In literature, Perry (Perry 1991) and Veenhuizen (Veenhuizen 1997) compared drug uptake after IV and IP injection. Perry (Perry 1991) described a longer tumoral elimination half-time (113.6 h vs 60.6 h) with IP administration, with a lower liver and kidney uptake. Veenhuizen (Veenhuizen 1997) described a higher disseminated tumoral drug uptake after IP administration too (about 20 times that after IV administration).

[0145] Effectiveness of PDT in colorectal peritoneal metastasis was knows for more than 30 years (Tochner 1985). Clinical trials showed important side effects (capillary leak syndrome and bowel perforation), mainly explained by low tumor-selectivity of the PS used (first generation mainly) (Pinto 2018). However, in preclinical studies, each new generation of PS permitted to improve tumoral targeting, with less toxicity and better effectiveness. Ascencio (Ascensio 2008), Estevez (Estevez 2010), Mroz (Mroz 2011), Hino (Hino 2013) and Kato (Kato 2017) described carcinomatosis necrosis, until 77% complete response (Ascensio 2008). Our results were inferior to those reported in the literature in 3 different PC models. Ascensio and Estevez [Ascencio 2008, Estevez 2010] found, 24 hours after PDT treatment, a mean necrosis score of 3.5. However, it was not the same animal model (rat vs mouse), not the same tumor model (ovary vs colon) and we have no information about the toxicity of the treatment on the other organs. However, Mroz [Mroz2011] described, after PDT in a mouse model of colorectal PC, a tumoral cytotoxicity by apoptosis and necrosis on the periphery of the nodules whereas it consisted rather in a necrosis in the center. In our study, necrosis and apoptosis concerned the both. Moreover, we first demonstrated the tumoral infiltration with macrophages and lymphocytes after PDT treatment.

[0146] Survival advantage of PDT was analyzed by some authors. Tochner [Tochner 1991], first, shown 85% survival at 25 days in PDT group, whereas all mice not receiving HPD-laser treatment died between days 20 and 23. Song et al [Song] shown PDT prolonged survival too. The median follow-up time was 45 days (95% CI, 1.17-88.83 days) in the treatment group versus 15 days (95% CI, 6.68-23; 32 days) and 19 days (95% CI, 13.16-24.84 days) in surgery alone group and surgery+laser without PS group (p=0.008). This prolonged survival was described by Yokoyama too (Yokoyama). Mean survival was compared in three groups: debulking surgery (DS) alone, DS+PDT and DS+PDT and clofibric acid. Survival was significantly longer 35.5 days, 46.3 days and 52.5 days respectively (p<0.005). We demonstrated the impact of PS vectorization on survival by comparing 2 vectors. EVs permitted a prolonged survival after PDT with 30% of survival in this group, whereas others treated mice were died. It notably allowed a superior survival compared to the liposomal formulation (p=0.005). These 2 vectorized formulations permitted to remove PDT lethal toxicity occurred after free mTHPC IP injection. Intraperitoneal vectorized PDT-related toxicity, such as bowel perforation, did not appear in our study. However, mice treated with non-vectorized PDT died 4 days following treatment. Histological analysis shown hepatic and kidney necrosis in this last group. Liver PDT toxicity is described. Tochner first, (Tochner 1991) shown in all treated animals hemosiderin-like deposits mainly in the periportal parenchymal cells and a very mild non-specific reactive hepatitis in which there was mild hyperplasia of Kuffer cells and few lymphohistiocytic aggregates. Perry (Perry 2001) published significant mortality in PDT groups and animals generally died from liver and small bowel necrosis. Griffin (Griffin 2001) didn't note major acute late clinical effects but all treated dogs and one control dog showed transient elevations in the LFTs, AlkP, AST, and ALT. These liver function test anomalies was noted by Guyon too (Guyon 2014).

[0147] To conclude, EVs permitted a biological nanovectorization of mTHPC with an important tumoral selectivity. In vivo studies proved that EVs-based PDT was effective for colorectal peritoneal metastasis. It permitted an intra-tumoral cytotoxic effect of PDT by direct and indirect mechanisms. Particularly, we observed an intra-tumor macrophages infiltration after PDT and a lymphocyte infiltration provided by the vesicles. This is the first time to our knowledge that this immunostimulatory effect is analyzed in vivo after vectorization of a PS in a CP model.

REFERENCES

[0148] Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure.