Julien Namur

Embolization of hepatocellular carcinoma with drug-eluting beads: Doxorubicin tissue concentration and distribution in patient liver explants

BACKGROUND & AIMS To follow the local tissue delivery of doxorubicin in HCC explants from patients embolized with drug-eluting beads and to compare it with histologic modifications. METHODS Six patients with HCC underwent chemoembolization with doxorubicin-eluting beads (caliber 100-300 μm, dose 75-150 mg) followed by liver transplantation at different time points (8 h to 36 days). On sections of the explanted liver, the tissue concentration of doxorubicin was determined radially around bead-occluded vessels with microspectrofluorimetry. The intra/peritumoral location of the beads and the modifications of the surrounding tissue were determined on an adjacent hematein-eosin-saffron-stained section and compared to drug measurements. RESULTS Doxorubicin was detected in the tissue surrounding the beads at all times of explantation. The drug impregnates an area of at least 1.2 mm in diameter around the occluded vessel. The tissue concentration of drug ranges from 5 μM at 8 h to 0.65 μM at 1 month. In patient transplanted at 8 h, no major tissue modification was observed and we found 42% of the beads occluding intratumoral vessels. Drug concentration was not different around intratumoral and peritumoral occluded vessels. After 9-14 days, necrosis was present around 37% of vessels and at 32-36 days, around 40% of vessels. Necrotic tissue was associated with a deeper penetration and a higher concentration of the drug than non necrotized areas, though statistically significant only at 32-36 days. CONCLUSIONS Doxorubicin-eluting beads provide a sustained delivery of drug for a period of 1 month and local tissue concentrations above cytotoxic threshold in HCC-bearing livers.

Drug-eluting Beads for Liver Embolization: Concentration of Doxorubicin in Tissue and in Beads in a Pig Model

PURPOSE To evaluate the local tissue concentrations of the antineoplastic agent doxorubicin and the amount of drug still present inside drug delivery embolization beads at different time points after embolization and to compare doxorubicin levels with histologic modifications around the beads in a pig liver model. It was hypothesized that doxorubicin-eluting beads maintain cytotoxic concentrations of drug locally over a period of several weeks, as suggested by in vitro elution tests. MATERIALS AND METHODS Left lobe hepatic artery embolization was performed in 10 pigs with 100-300-microm or 700-900-microm beads loaded with 37.5 mg doxorubicin/mL. Control unloaded 100-300-microm beads were injected in five pigs. Livers were sampled 28 days or 90 days after embolization. The amount of drug retained inside the beads was assessed with infrared microspectroscopy. Doxorubicin concentration and distribution in the tissue around the beads were determined with microspectrofluorimetry and compared with tissue modifications on hematein eosin saffron-stained sections. RESULTS Doxorubicin-eluting beads eluted 43% of their initial drug load after 28 days and 89% after 90 days. Doxorubicin was present in tissues around the beads at both time points, with a significant decrease over time (P = .0004). The drug was detected at distances as far as 600 microm from the bead edge. Doxorubicin tissue concentrations ranged from 0.55 microM to 6.80 microM, [corrected] which are cytotoxic levels in hepatocyte cell cultures. High concentrations of drug were associated with coagulative necrosis of liver parenchyma. Doxorubicin-eluting beads 100-300 microm in size induced more necrosis than 700-900-microm beads (P = .0036). CONCLUSIONS Doxorubicin-eluting beads deliver high concentrations of the drug over a period of at least 3 months at several hundred micrometers from the bead, leading to significant cytotoxic effects.

Arterial distribution of calibrated tris-acryl gelatin and polyvinyl alcohol microspheres in a sheep kidney model.

Objective:The objective of this study was to compare the repartition in the renal arterial vasculature of tris-acryl gelatin microspheres (TGMS) and polyvinyl alcohol microspheres (PVAMS) of 3 calibers (500–700, 700–900, and 900–1200 μm). Materials and Methods:Twelve kidneys from 6 adult sheep were embolized and histologically analyzed. The number and size of microspheres and vessels were measured, as well as the deformation of TGMS and PVAMS, and the histologic location according to a classification in 5 zones of the kidney. Results:Two hundred eighty-four vessels were measured. The diameter of the occluded vessels increased when the caliber used for embolization was larger for TGMS and for PVAMS (P < 0.0001, each). The location of TGMS and PVAMS within the vasculature was different for each caliber, because PVAMS blocked significantly more distally than TGMS (P < 0.0001 each). The deformation within the tissue was greater for PVAMS (18.0 ± 12.3%) than for TGMS (9.0 ± 8.3%) in general (P < 0.001) and for each caliber of injected microspheres (P < 0.001 each). Conclusion:The repartition of a spherical embolic agent in a given vascular network can be influenced by its size and also by its deformation within the vascular bed.

Modified Model of VX2 Tumor Overexpressing Vascular Endothelial Growth Factor

PURPOSE To determine whether upregulated expression of vascular endothelial growth factor (VEGF) in VX2 cells can increase vessel density (VD) and reduce tumor necrosis. MATERIALS AND METHODS The VX2 cell line was transfected with expression vectors containing cDNA for rabbit VEGF. Stable clones producing rabbit VEGF (VEGF-VX2) were selected. VEGF-VX2 cells (n = 5 rabbits) or nontransfected VX2 cells (controls; n = 5 rabbits) were implanted into leg muscle of 10 rabbits. The animals were sacrificed at day 21. Tumor volume, percentage of necrosis, VD, and VEGF concentration in tumor protein extract were quantified. RESULTS Overexpression of VEGF by VX2 cells augmented tumor implantation efficiency 100% and favored cyst formation. The tumor volume was significantly larger for VEGF-VX2 transfected tumors versus controls (P = .0143). Overexpression of VEGF in VX2 cells significantly increased the VD of the tumors (P = .0138). The percentage of necrosis was reduced in VEGF-VX2 tumors versus controls (19.5% vs 38.5 %; P = .002). VEGF concentration in VEGF-VX2 tumors was significantly higher than in control tumors (P = .041) and was correlated with tumor volume (ρ = .883, P = .012). CONCLUSIONS The overexpression of VEGF increased tumor growth and vascularization, favored cyst formation, and reduced tumor necrosis. This new phenotype of the VX2 tumor may offer some advantages over classic models of VX2 tumor for evaluating anticancer therapies.

IL6 and TNF expression in vessels and surrounding tissues after embolization with ibuprofen-loaded beads confirms diffusion of ibuprofen

PURPOSE In the treatment of uterine fibroid embolization related pain, the use of embolics loaded with non-steroidal anti-inflammatory drugs (NSAID) relies on an efficient delivery and impregnation of the embolized tissue. Immuno-labelling and spectroscopic techniques have demonstrated the release of ibuprofen from drug eluting beads (Wassef et al., 2008; Namur et al., 2009) but failed to demonstrate diffusion of the drug beyond the vascular wall (VW). We investigated whether ibuprofen diffused beyond the VW in surrounding tissues (ST), by tracking its biological effects through the modulation of expression of two main inflammatory cytokines. MATERIALS AND METHODS Uterine arteries of 6 sheep were embolized with ibuprofen loaded beads (IBU-BB) or non-loaded beads (BB) and sacrificed at one week. On frozen tissue slices, VWs of occluded arteries were isolated from ST using laser capture microdissection. RNA was extracted from VW and ST samples. Gene expression of IL6 and TNFα genes was measured by quantitative real-time PCR (qPCR). RESULTS IL6 expression was significantly increased in IBU-BB compared to BB group both in VW (VW: fold-change (FC)=4.9, p=0.0009) and ST (ST: FC=8.7, p=0.0003). In IBU-BB, IL6 was significantly more expressed in VW than in ST (FC=4.4; p=0.0009). TNFα expression was not significantly different between IBU-BB and BB groups. CONCLUSION Using qPCR+microdissection was useful to evaluate the spread of the biological effects of drug-loaded systems which attest of the tissular release. This approach can be considered when other drug detection techniques are unsuccessful or difficult to achieve. IL6 can be used as a marker of ibuprofen released by drug eluting beads in uterus. Gradient of expression of IL6 suggests diffusion of ibuprofen across the VW into the ST.

Automated Quantification of Tumor Viability in a Rabbit Liver Tumor Model after Chemoembolization Using Infrared Imaging

The rabbit VX2 tumor is a fast-growing carcinoma model commonly used to study new therapeutic devices, such as catheter-based therapies for patients with inoperable hepatocellular carcinoma. The evaluation of tumor viability after such locoregional therapies is essential to directing hepatocellular carcinoma management. We used infrared microspectroscopy for the automatic characterization and quantification of the VX2 liver tumor viability after drug-eluting beads transarterial chemoembolization (DEB-TACE). The protocol consisted of K-means clustering followed by principal component analysis (PCA) and linear discriminant analysis (LDA). The K-means clustering was used to classify the spectra from the infrared images of control or treated tumors and to build a database of many tissue spectra. On the basis of this reference library, the PCA-LDA analysis was used to build a predictive model to identify and quantify automatically tumor viability on unknown tissue sections. For the DEB group, the LDA model determined that the surface of tumor necrosis represented 91.6% ± 8.9% (control group: 33.1% ± 19.6%; Mann-Whitney P = 0.0004) and the viable tumor 2.6% ± 4% (control group: 62.2% ± 15.2%; Mann-Whitney P = 0.0004). Tissue quantification measurements correlated well with tumor necrosis (r = 0.827, P < 0.0001) and viable tumor (r = 0.840, P < 0.0001). Infrared imaging and PCA-LDA analysis could be helpful for easily assessing tumor viability.

Safety and efficacy compared between irinotecan-loaded microspheres HepaSphere and DC bead in a model of VX2 liver metastases in the rabbit.

PURPOSE To compare irinotecan-eluting HepaSphere (BioSphere Medical, Roissy-en-France, France) and DC Bead (Biocompatibles UK Ltd, London, United Kingdom) embolization microspheres for distribution in tumors, release properties, tolerance, and antitumor effects in a model of liver metastases in the rabbit. MATERIALS AND METHODS Multiple liver tumors were created by injection of a VX2 cell suspension in the portal vein of rabbits. After 2 weeks, embolization of the proper hepatic artery was performed with a fixed volume of bland HepaSphere (n = 5), irinotecan-loaded HepaSphere (n = 6), or irinotecan-loaded DC Bead (n = 5) microspheres. Untreated animals injected with VX2 cells served as control animals (n = 5). Plasma pharmacokinetics of irinotecan and its metabolite SN38 were assessed. Histopathology and gene expression analysis were performed 3 days after treatment. RESULTS Among all treated groups, there was no significant difference in liver enzymes or liver damage on histology. Irinotecan-loaded HepaSphere microspheres showed a faster release of drug than DC Bead microspheres leading to a twofold higher concentration of drug in plasma for HepaSphere microspheres. HepaSphere microspheres were less frequently found inside tumor nodules on histology than DC Bead microspheres (11% vs 48%, P < .001) because of their larger size. Tumor necrosis was significantly greater for rabbits given irinotecan-loaded HepaSphere microspheres (69% of total tumor surface) and rabbits given DC Bead microspheres (50% of total tumor surface) compared with control animals (24% of total tumor surface, P = .006 and P = .047). CONCLUSIONS HepaSphere and DC Bead microspheres loaded with irinotecan caused significant necrosis of tumor nodules in a model of VX2 liver metastases. This outcome was mostly due to irinotecan delivery rather than vascular occlusion by the microspheres and was greater for HepaSphere microspheres compared with DC Bead microspheres.

Laparoscopic subperitoneal injection of chemo-loaded particles lowers tumor growth on a rabbit model of peritoneal carcinomatosis

The purpose of our study was to assess the effect of controlled-release chemotherapy on the growth and viability of peritoneal carcinomatosis treated by subperitoneal injection in a rabbit VX2 model. A model of peritoneal carcinomatosis was created by laparoscopic injection of VX2 tumor in the left and right broad ligaments of 12 White New Zealand rabbits. At day 12, each tumor was randomly treated with a peritumoral injection of 0.5 mL microspheres loaded with doxorubicin (DEM-DOX) or unloaded (DEM-BLAND). Seven days after treatment, tumor volume, tumor viability in histology, local tumor necrosis in contact with DEM, and doxorubicin concentration profile around the drug eluting microspheres (DEM) were measured. Tumor volume was significantly lower in the DEM-DOX group (3.6 ± 3.2 cm3) compared with the DEM-BLAND group (8.9 ± 5.4 cm3) (p = 0.0425). The percentage of viable tumor tissue was significantly lower in the DEM-DOX group (38% ± 17%) compared with the DEM-BLAND group (56% ± 20%) (p = 0.0202). Tissue necrosis was observed around all DEM-DOX up to a distance of 1.094 ± 0.852 mm and never observed around DEM-BLAND. Drug concentration was above the therapeutic level of 1.0 µM up to a distance of 1.4 mm from the DEM to the tumor. Laparoscopic subperitoneal injection of chemo-loaded particles is feasible and lowers tumor growth and viability in a rabbit model of peritoneal carcinomatosis after 1 week.