ORIGINAL RESEARCH published: 20 August 2020 doi: 10.3389/fmed.2020.00415 Development and Characterization of a New Endoscopic Drug-Eluting Platform With Proven Efficacy in Acute and Chronic Experimental Colitis Ignacio Bon 1 , Mary Cano-Sarabia 2 , Napoleon de la Ossa 3 , Ramon Bartolí 1,4*† and Vicente Lorenzo-Zúñiga 1,5*† 1 Health Research Institute Germans Trias i Pujol (IGTP), Barcelona, Spain, 2 Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC and The Barcelona Institute of Science and Technology, Campus de la UAB, Bellaterra, Spain, 3 Servicio de Anatomia Patológica, Hospital Universitari General de Catalunya—Grupo Quirón Salud, Barcelona, Spain, Edited by: 4 Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBEREHD), Barcelona, Spain, Michel Goldman, 5 Endoscopy Unit, University Hospital La Fe, Valencia, Spain Institute for Interdisciplinary Innovation in Healthcare (I3h), Belgium Reviewed by: Background and Aims: Mucosal lesions refractory to biological treatments represent Juan Ricardo Marquez, unmet needs in patients with inflammatory bowel disease (IBD) that require new treatment Instituto de Coloproctología ICO S.A.S., Colombia modalities. We developed and characterized a new endoscopic drug-eluting hydrogel Angel Lanas, (CoverGel) with proven efficacy in acute and chronic experimental colitis (EC) in rats. University of Zaragoza, Spain *Correspondence: Methods: CoverGel was developed based on appropriate rheological, drug release, Ramon Bartolí gelation, structural, and degradation property capacities to allow endoscopic application. rbartolisole@gmail.com Experimental colitis (EC) was induced by TNBS application in rats. In acute EC 40, Vicente Lorenzo-Zúñiga vlorenzozuniga@gmail.com rats were randomized in five groups (eight each): Sham, Control, CoverGel, CoverGel † These + Infliximab (IFX) and CoverGel + Vedolizumab (VDZ). In chronic EC, 12 rats were authors have contributed equally to this work randomized in two groups (six each): IFX s.c. and CoverGel + IFX. Endoscopic, histological, and blood test were performed during follow-up to evaluate clinical success. Specialty section: Antibodies to IFX (ATIs) were evaluated in chronic EC animal study. This article was submitted to Gastroenterology, Results: CoverGel is a biocompatible and bioadhesive reverse thermosensitive gelation a section of the journal hydrogel with a macroporous structure and drug release capacity. In acute EC animals Frontiers in Medicine treated with CoverGel + IFX or CoverGel + VDZ showed significantly clinical success Received: 08 January 2020 Accepted: 29 June 2020 (weight recovery, mucosal restoration, and bacterial translocation) as compared with Published: 20 August 2020 controls and animals without a bioactive drug. In a chronic EC animal study, clinical Citation: efficacy was comparable in both groups. Levels of ATIs were significantly lower in animals Bon I, Cano-Sarabia M, de la Ossa N, Bartolí R and Lorenzo-Zúñiga V (2020) treated with CoverGel + IFX vs. IFX s.c. (0.90 ± 0.06 µg/mL-c vs. 1.97 ± 0.66 µg/mL-c, Development and Characterization of p = 0.0025). a New Endoscopic Drug-Eluting Platform With Proven Efficacy in Acute Conclusions: CoverGel is an endoscopic vehicle to locally deliver biological drugs with and Chronic Experimental Colitis. proven efficacy in acute and chronic EC in rats and induce less immunogenicity reaction. Front. Med. 7:415. doi: 10.3389/fmed.2020.00415 Keywords: drug-eluting, experimental colitis, endoscopic shielding technique, hydrogel, endoscopy Frontiers in Medicine | www.frontiersin.org 1 August 2020 | Volume 7 | Article 415 Bon et al. New Endoscopic Drug Eluting Platform INTRODUCTION into the gel and redrawn. The velocity of the insertion assay was 1 mm/s, and the withdrawn distance was 9 mm. Adhesion was Interventional inflammatory bowel disease (IBD) endoscopy has determined at 20 and 37◦ C, and the (iii) microstructure of the an expanding role in the era of biologic therapy. A significant hydrogel was assessed with scanning electron microscopy (SEM). number of patients have mucosal lesions refractory to biological Gelation assay was carried out to determine the behavior of treatments despite the availability and growing use of anti- the hydrogel under UV light radiation at different temperatures. tumor necrosis factor (TNF), anti-integrin, anti-interleukin, and The following scenarios were tested: (i) 1 mL of hydrogel and 15 new small-molecule (1, 2). These drugs have changed the way µL of photo-initiator solution were mixed and placed on a petri of treating IBD patients, but those situations represent unmet dish. The hydrogel was heated at 37◦ C for 30 min, and gelation needs that require new treatment modalities to prolong the was tested. Next, the hydrogel was cooled; (ii) 1 mL of hydrogel intervention-free period. An endoscopic shielding technique and 15 µL of photo-initiator solution were mixed and placed on with hydrogels has been proposed by our group as a less invasive a petri dish. The dish was placed under ultraviolet light 2,000 approach to reduce or postpone the need for surgical resection mV/cm2 for 3 min (AnalytikJeta UVP Crosslinker), keeping the (3, 4). temperature below 20◦ C. The hydrogel was heated at 37◦ C for Hydrogels or hydrophilic gels are cross-linked polymeric 30 min, and gelation was tested. Next, the hydrogel was cooled; three-dimensional networks which present soft tissue-like elastic, (iii) 1 mL of hydrogel was placed on a petri dish. The hydrogel non-toxic, and biodegradable properties while being responsive was heated at 37◦ C for 30 min, and gelation was tested. Next, to stimuli (5–8). Nowadays we do not have any valid endoscopic the hydrogel was cooled; (iv) 1 mL of hydrogel was placed on treatment to IBD mucosal lesions in patients with refractory a petri dish. The dish was placed under ultraviolet light 2,000 or loss of response to biological therapy. The purpose of the mV/cm2 for 3 min, keeping the temperature below 20◦ C. The present study was to develop and characterize a new endoscopic hydrogel was heated at 37◦ C for 30 min, and gelation was tested. drug-eluting platform and test its efficacy in acute and chronic Next, the hydrogel was cooled. For hydrogel observation after the experimental colitis in rats. irreversible episode, they were lyophilized at 50◦ C for 96 h and were examined by Field-Emission Scanning Electron Microscopy MATERIALS AND METHODS (FE-SEM) on a FEI Quanta 650 microscope using aluminum tape as support. Characterization of the Drug-Eluting Degradation assay was performed to understand the Platform degradation rate of our hydrogel under different conditions We have developed a new hydrogel (CoverGel) comprising that would try to mimic the different conditions found in the specific amounts of hyaluronic acid (HA), methylcellulose (MC), gastrointestinal tract; 1 mL of hydrogel was irreversibly gelified and Poloxamer Pluronic-F127 (F127). This platform has been and then submerged in 10 mL of one of the following solutions: patented (WO 2018/019881 A1). The composition of CoverGel PBS pH = 7, PBS pH = 3, PBS pH = 9, PBS pH = 11 and rat feces is based on the combination of different substances to create this bacteria culture on Luria-Bertani (LB), and broth at a starting solution, which can be prepared before the procedure. range of 105 CFU/mL. The piece of gelified hydrogel was weighed Our goal was to improve the beneficial effects of HA before the submersion (Wo) and then on set date points (Wt) onto the colonic mucosa by adding different substances to until complete dissolution of the hydrogel or until 1 month under increase biocompatibility, bioadhesive, and bioactive properties the conditions was reached. At each time point, the medium was and make it suitable to be administered through the endoscope’s also changed when the weighing took place. working channel. For this, we evaluated different aspects: Drug absorbance and drug release kinetics of the hydrogel rheological properties (viscosity, adherence, and microstructure), was also evaluated in vitro with mathematical models to study gelation, degradation, drug absorbance, drug release, and the capacity of our hydrogel to both release and absorb molecules biocompatibility. Four different original combinations (colon of different molecular weight, in order to further study the dressing, CD) of prototypes of materials (CD1, CD2, CD3, CD4) capability of this hydrogel to act as a drug-eluting platform. For were tested for viscosity, adhesion force, and microstructure. these experiments, we worked with two molecules: (i) Trypan The best composition for the drug-eluting platform was different blue, which has a molar mass of ∼ 960 Daltons; and (ii) bovine percentages of HA, MC, and PF127. All compounds of the serum albumin (BSA), which has a molecular weight of 66.5 kDa. product are well-known and have proven safety. The composition For the Trypan blue experiments, first, a calibration curve was of the invention is biodegradable, is easy to administrate, remains prepared by measuring the absorbance (λ = 450 nm) of different adhered to the mucosal lesion, and is a drug-release composition. concentrations of Trypan blue diluted in water using a Uv-Vis Rheological test: (i) Viscosity was evaluated in a rheometer spectrophotometer (Varioskan Flash, Thermo). SkanIt Software (Haake RheoStress; Thermo Fisher Scientific, Waltham, MA, 2.4.1 was used to analyze results. Then, the curve equation was USA) with a C60/1◦ Ti probe and a gap set of 0.053 mm. The obtained: y = 6234.9x + 0.1008. R2 = 0.99349. Absorbance rotation ramp went from 0–300 s−1 in 30 s. For each study, and release of BSA was evaluated with the Bicinchoninic acid Viscosity (η) was measured as a function of shear rate (γ) at 22◦ C (BCA) Protein Assay Kit (PierceTM , Thermo Scientific) following (room temperature) and 37◦ C (body temperature), (ii) Adhesion manufacturer’s instructions. was measured with a Texture Analyzer TA.XT Plus (Stable Micro To incorporate substances inside the hydrogel, 0.25 mL of Systems, Surrey, UK). A 40-mm (diameter) disk was compressed gelated hydrogel was submerged in 5 mL of 0.0005 mg/mL Frontiers in Medicine | www.frontiersin.org 2 August 2020 | Volume 7 | Article 415 Bon et al. New Endoscopic Drug Eluting Platform Trypan blue or 2 mg/mL BSA. Hydrogels incorporated the for 24, 48, 72, and 96 h. At specific time points, cell viability was solution at room temperature for 48 h. To calculate the amount evaluated using Resazurin Sodium Salt (Sigma-Aldrich). of substance incorporated into the matrix of the hydrogel, the Hemolysis test was done in vitro. Two milliliters of gelified following equation was used: hydrogel were placed in 20 mL of sterile saline and incubated for 72 h at 37◦ C. Afterward, the solution was collected and V1 • C1 − V2 • C2 filtered with a 0.22-µm membrane; 2-mL blood samples were Incorporating efficiency (%) = × 100 V1 • C1 freshly collected from six male Sprague–Dawley rats into an anticoagulant tube and gently mixed. Blood was then diluted with Where V1 is the original volume of the substance solution (mL); 2.5 mL of saline. For the hemolysis test, 10 mL of the hydrogel V2, the remaining volume of the substance solution after 48 h extraction, distilled water, and saline were poured in 50 mL (mL); C1, the initial concentration of the substance (µg/mL); and test tubes and placed at 37◦ C for 30 min. After that, 0.2 mL of C2, the remaining concentration of the substance (µg/mL). diluted blood was added to each test tube with gentle shaking, Hydrogels were withdrawn from the solution and placed in and the solution was placed at 37◦ C for 1 h. The absorbance of a release medium (PBS). At specific time points, the amount of the samples after 1 h was evaluated with a spectrophotometer substance released into the medium was evaluated as mentioned (Varioskan Flash, Thermo) at 545 nm. SkanIt Software 2.4.1 was before. The kinetics of drug release from the hydrogels was used to analyze results. Hemolysis ratio (%) was calculated with studied with three different models to asses which one fitted best the following formula: with the release data from our experiments. The three different models were the following: A hydrogel − A saline solution × 100 A distilled water − A saline solution Higuchi release model : Mt = Kh t1/2 Zero − order release model : Mt = K0 t The acute toxicity of CoverGel was evaluated in male Sprague– First − order release model : Mt = 1 − e− K1 t Dawley rats, 7 months of age (400–550 g). Animals were divided in three groups: two subjects and a control group. Each group Where Mt is the fraction of drug released at each time point contained three animals, and a total of nine animals were (t), and Kh , K0 , and K1 are the Higuchi release kinetic constant, used. Rats from the subject groups were injected with 10 the zero-order release kinetic constant, and the first-order release mL/kg of our hydrogel into the abdominal cavity, and rats kinetic constant, respectively. from the control group were injected with 10 mL/kg saline Lastly, the drug release mechanism was analyzed using the into the abdominal cavity. Animals were observed daily after Korsmeyer-Peppas model: administration and evaluation of general conditions (activity,   hair, feces, behavior, and other clinical signs), body weight, Mt and mortality were done. Animals were euthanized at 3 and 7 F= = Km t n M days with anesthetic overdose, and major organs (heart, liver, spleen, lung, and kidney) were extracted, evaluated, and fixed This is a semi-empirical model used when the exact mechanism in a 10% formaldehyde solution. Organs were then stained is not known or in the case when more than one mechanism with hematoxylin/eosin for histopathological evaluation. Blood is involved in the drug release; where F is the fraction of drug was also extracted, and hematological and biochemistry analyses released at time (t), Mt is the amount of drug released at time (t), were performed. M is the total amount of drug in the hydrogel, Km is a kinetic The effect of subcutaneous implantation of CoverGel was constant, and n is the release exponent, indicative of the drug evaluated in six male Sprague–Dawley rats, 7 months of age (420– release mechanism. n is estimated from linear regression of log 500 g). Rats maintained the subcutaneous implantation for 10 (Mt/M) vs. log t; when determining the n exponent, only portions days. Under anesthesia, a 2-cm incision was done on the dorsal of the release curve ≤ 60% should be used. back of the rat, and 0.5 g of gelified hydrogel was implanted Biocompatibility assays (cytotoxicity, hemolysis, and acute subcutaneously. The incision was then sutured, and rats were toxicity) were first performed to check the possible cytotoxicity allowed to recover with food and water ad libitum. After 10 days, induced by CoverGel, carrying out a co-culture of a cell rats were euthanized, and the hydrogel was extracted and cut line with different conditions involving our hydrogel. The in two; one piece was placed in Carnoy solution (60% ethanol, cell line chosen for these experiments was Caco-2 (ATCC R 30% chloroform, and 10% glacial acetic acid) for histological HTB-37TM), a human colorectal adenocarcinoma cell line. For evaluation and the other piece was lyophilized (Christ Loc-1 m, cytotoxicity evaluation, 3 × 104 cells/cm2 were cultured in 24- B. Braun) and evaluated using SEM. well cell plates under the following conditions: Normal culture medium (DMEM medium supplemented with 10% fetal bovine Endoscopy serum, 1% L-glutamine, and 1% penicillin/streptomycin) as a All endoscopic evaluations were carried out using an Olympus positive control; Normal culture medium supplemented with Video bronchoscope EVIS EXERA II (BF-1T180)-type 10% dimethyl sulfoxide (DMSO) as a negative control; and endoscope with an outer diameter of 6.0 mm and a working Normal culture medium supplemented with 1%-2%-5%-10%- channel diameter of 3.0 mm. Room air was used for insufflation 15% of hydrogel v/v. Cells were cultured at 37◦ C and 5% CO2 during the endoscopy. Endoscopic application of hydrogel Frontiers in Medicine | www.frontiersin.org 3 August 2020 | Volume 7 | Article 415 Bon et al. New Endoscopic Drug Eluting Platform was done by positioning the tip of the catheter over the in the acute EC study. Animals underwent endoscopic follow- mucosal lesions. up at days of induction, days of treatment, and the day of euthanasia. Furthermore, antibodies to IFX (ATIs) in serum were Acute Experimental Colitis (EC) Animal also evaluated. To separate ATIs from IFX as they might be found on blood, a protein G column and an acidic buffer treatment Study were used; 600 µL of a 1:10 dilution of serum was applied The EC animal model was induced by the rectal instillation onto the protein G column (Ab Spin trap; GE Healthcare) and of 0.6 mL of 3.5% TNBS solution in 50% ethanol (Day 0). washed. Trapped ATIs were eluted with 400 µL of elution buffer The peak of colitis was established 3 days post administration (0.1 M Glycine-HCl, pH 2.7) which releases proteins trapped in (Day 3), and animals were examined to evaluate the affection the protein G column and, at the same time, dissociates ATI’s- and apply treatments. Forty male Sprague–Dawley rats (Harlan IFX complexes. Samples were eluted into neutralizing buffer (1M Laboratories Models SL, Barcelona, Spain) weighing 250–300 g, Tris-HCl pH 9) containing an IgG concentration of 20 µg/mL to which lost at least 8% of the original weight (Day 0) and showed prevent the reformation of immune complexes. Then, with the a circumferentially affected area at day 3, were included in this samples obtained from the protein G column, a 96-well plate study. Animals were randomized in five groups: (i) Sham group, was coated with 200 µL/well O/N at 4◦ C or for 1 h at 37◦ C. healthy animals with no administration of TNBS; (ii) Control Plates were washed with washing buffer (PBS 0.1% Triton), and Group, animals with EC treated with 1 mL of saline, through the 200 µL/well horseradish peroxidase-conjugated Infliximab (HRP endoscope; (iii) CoverGel group, animals with EC treated with conjugation Kit, Abcam) (0.2 µg/mL IFX-HRP) was added for CoverGel without active substance; (iv) CoverGel + Infliximab 30 min at room temperature, protecting the plate from direct (IFX) group, animals with EC treated with CoverGel with IFX, light. The plate was further washed, and finally, a substrate 1 mg/mL; (v) CoverGel + Vedolizumab (VDZ) group, animals solution (3,3′ ,5,5′ -Tetramethylbenzidine/H2 O2 50% vol/vol) was with EC treated with CoverGel with VDZ, 1 mg/mL. A total added to the microplate. The enzyme reaction yields a blue amount of 1.5 mL of CoverGel was applied in groups iii, iv, and v; product that turns yellow when stop solution (2 M sulfuric acid) thus, the dosage of the drug applied to groups iv and v was 1.5 mg is added. The intensity of the color measured is in proportion of IFX or VDZ, respectively. of the quantity of ATIs bound in the initial step; this intensity is Animals were weighed and underwent colonoscopy on days 0, measured by determining the optical density of each well using a 3, and 7. Blood was extracted by a small cut on the tail, and serum microplate reader at 450 nm with a wavelength correction set to was kept frozen for future evaluations. Ponderal evolution was 540 nm using a spectrophotometer (Varioskan Flash, Thermo). studied as the variation (1) of original weight at day 0 for each SkanIt Software 2.4.1 was used to analyze results. rat. Four days after treatment (Day 7), animals were euthanatized by an anesthetic overdose, and liver and colon were evaluated. Ulcer site samples were cut and placed in formaldehyde for H/E Statistical Study staining and blinded histopathological evaluation. Histological All values are expressed in median ± range unless otherwise study evaluated the maintenance of the intestinal architecture stated. The comparison of groups was done using two-way and the inflammatory cell infiltrate in a 0–3 scale for both ANOVA and Tukey’s multiple comparison test as a post hoc. characteristics and a final 0–6 scale. The scoring system (Table 1) Release kinetics was studied using mean values of 6 different followed was extracted from Ulrike Erben et al. (9). Liver was experiments and a non-linear regression fit to each different retrieved under sterile conditions and 0.5 g was places in 1 mL model equation. The Graphpad Prism 6.0 software was used to of saline and homogenized. Then, 0.1 mL of the homogenate perform analysis. was cultured on a LB Agar petri dish at 37◦ C for 24 h. The appearance of CFU was evaluated as a way to establish bacterial translocation to the liver, a marker, or a weak intestinal wall due Animal Care to inflammation. All animals were obtained from ENVIGO. All protocols were reviewed by the Animal Experimentation Ethics Committee of Chronic EC Animal Study the Hospital Universitari Germans Trias I Pujol (registered as All animals included in this protocol underwent four rounds of B9900005) and approved by the Departament d’Agricultura, TNBS colitis induction in the same fashion as the acute model, Ramaderia, Pesca, Alimentació i Medi Natural of the Catalan thus, induction was done on days 0, 7, 14, and 21; treatments were Regional Government (protocols number 8,585 and 10,235) given on days 3, 10, 17, and 24, and euthanasia was performed according to current national and European Union legislation on day 28. Twelve male Sprague–Dawley rats were included. regarding the protection of experimental animals. Rats were Animals were randomized in two groups: (i) IFX s.c. group, supervised daily in order to ensure animal welfare and animals with EC treated with 5 mg/kg of IFX subcutaneously; (ii) euthanized, if required, by anesthetization with isoflurane. All CoverGel + IFX group, animals with EC treated with CoverGel animals had free access to water and food ad libitum. After any + IFX (1 mg/mL). A total amount of 1.5 mL of CoverGel was procedure carried on the animals, 200 mg/kg of paracetamol applied per animal; thus, the total amount of drug administered was administered intraperitoneally, and animals were monitored to group ii was 1.5 mg of IFX per animal at each day of daily. Normal parameters of well-being such as hair health and treatment. All assessments were carried out the same way as secretions were analyzed. Frontiers in Medicine | www.frontiersin.org 4 August 2020 | Volume 7 | Article 415 Bon et al. New Endoscopic Drug Eluting Platform TABLE 1 | Histological scoring system for the experimental colitis animal models. Inflammatory cell infiltrate Intestinal architecture Severity Extent Score 1 Epithelial changes Mucosal architecture Score 2 Mild Mucosa 1 Focal erosions 1 Moderate Mucosa and submucosa 2 Erosion ± Focal ulcerations 2 Marked Transmural 3 Extended ulcerations ± granulation tissue ± pseudopolyps 3 Sum of scores 1 and 2: 0–6 FIGURE 1 | Rheological evaluations of the four original colon dressings (CDs). (A) Adhesion force evaluated at 22◦ C and 37◦ C expressed in mN/s. (B) Viscosity evaluated at 22◦ C (black line) and 37◦ C (red line) expressed in pascals. (C) Scanning electron microscopy images of CD1, CD3, and CD4 (top three pictures) and CD2 (bottom pictures). RESULTS Reversible and irreversible gelation of the hydrogel was acquired. Hydrogel can be given smoothly through a catheter Development of the Drug-Eluting Platform (diameter 2.0–2.2 mm) by applying a force of 370 mmHg All hydrogels showed a similar adhesion force at 22◦ C, but (0.48 atmospheres). Hydrogel was also stable during 30 days when temperature was raised, adhesion force of only CD2 at physiological-like conditions (Figure 2A). For Trypan blue, (CoverGel) was highly increased from −25 to −3,993 mN/s the incorporating efficacy was 58.96 ± 2.8%, and for BSA, (Figure 1A). Viscosity (η) of all samples was similar at 22◦ C, 24.09 ± 4.34% (mean ± SD). Drug release from hydrogel was but when temperature was increased to 37◦ C, the viscosity of related to the concentration of the drug in the medium, which CD2 was increased from < 1 Pa·s at 22◦ C to 1,500 Pa·s at 37◦ C followed a Fickian diffusion. The release kinetics model that (Figure 1B). In addition, the morphology of irreversible hydrogel best fits our results is the Higuchi release model, being the was evaluated by SEM observing that it is formed by a structured final equation obtained for our results: for Trypan blue (Mt = network with well-defined porous distribution ranging from 50 0.05760t1/2 ; R2 = 0.8986) and for BSA (Mt = 0.1369t1/2 ; R2 = to 300 µm (Figure 1C). 0.9167). The release exponents (n) were 0.4338 for Trypan blue Frontiers in Medicine | www.frontiersin.org 5 August 2020 | Volume 7 | Article 415 Bon et al. New Endoscopic Drug Eluting Platform Macroscopic evaluation of major organs didn’t show any signs of abnormalities, and organ weight with respect to total body mass didn’t present any significant differences (Figure 4A). Heart tissue didn’t show signs of toxicity; cardiac myocytes maintain a good arrangement; and no hemorrhage, inflammation or necrosis is observed. In the liver, no hepatocellular degeneration occurs, and there isn’t an abnormal neutrophil, lymphocyte, or macrophage infiltrate in the tissue. Tissue structure from spleen, kidneys, and lungs is maintained as well (Figure 4B). Subcutaneous placement of 0.5 g of the hydrogel showed a synovial metaplasia, a capsular surface covered by fibrohistiocytic cells that are grouped forming pseudovilli. No toxicity or inflammation was seen in the surrounding tissue, meaning that the body responded by encapsulating the hydrogel as it would do with a foreign body, but no rejection response was triggered (Figure 4C). At the time of placement, hydrogels weighed 0.5 g; when extracted, the weight was 0.95 ± 0.06 g. The platform was able to maintain its porosity and inner structure. No cell infiltration was seen 10 days after subcutaneous placement (Figure 4D). Efficacy of the Drug-Eluting Platform in Acute EC Animal Study At day 3, all animals presented an entire circumferential affectation of the wall. Macroscopic and endoscopic evaluations at day 7 showed a complete mucosal restoration (vascular pattern, ulceration and friability) in the CoverGel + IFX and CoverGel + VDZ groups, whereas in the CoverGel group, this mucosal FIGURE 2 | Characterization of CoverGel. (A) Degradation of the hydrogel restoration was not complete. However, all treated groups clearly when placed for 30 days under different conditions ( PBS pH3, PBS showed less mucosal damage than the Control group (Figure 5). pH7, PBS pH9, PBS pH 11, and bacterial culture). Graph represents weight of hydrogel over days. (B) Drug-release kinetics for All animals with acute EC had lost more than 8% of original Bovine serum albumin and Trypan blue. Graph represents concentration weight at day 3. At day 7, control animals showed a significant of substance on release medium (Mt) over time. Lines represent the different weight loss (−55%) as compared with the other groups. The mathematical models tested. drug-eluting platform with IFX or VDZ showed a significant weight recovery in comparison with CoverGel without bioactive TABLE 2 | Coefficient of determination (R2 ) of our data to the three kinetic release drug (−17 and −20 vs. −36%; p < 0.044) (Figure 6A). BT to models (zero-order release model, first-order release model, and Higuchi release liver at day 7 was absent in sham animals. The Control group model) and for the drug release mechanism (Korsmeyer-Peppas model). presented a significantly higher BT as compared with CoverGel + IFX and CoverGel + VDZ (100 vs. 14.3 and 16.6%; p < 0.05). Sample 0-order R2 1st-order R2 Higuchi R2 Korsmeyer-Peppas Animals treated with CoverGel showed a decreased BT (62.5%) Tryp. B 0.2022 0.3361 0.8986 0.9169 without statistical differences (Figure 6B). The histological score BSA 0.5393 0.8324 0.9167 0.9171 for groups with EC was comparable: Control 5.87 ± 0.35, CoverGel 5.24 ± 0.76, CoverGel + IFX 5.34 ± 0.98, and CoverGel + VDZ 5.28 ± 0.82 (Figure 6C). The ratio between colon length and weight [median (range)], a simple indicator of and 0.5116 for BSA (Figure 2B). Table 2 shows the correlation edema and inflammation in the tissue, was significantly higher coefficient of the different kinetic models and the Korsmeyer- in control animals as compared with the other treatment groups Peppas mechanism model. (Control vs. CoverGel, CoverGel + IFX and CoverGel + VDZ) CoverGel showed good biocompatibility, both in the in [0.3157 (0.244) g/cm vs. 0.2063 (0.1012) g/cm p = 0.0012, 0.1854 vitro and in vivo studies. Caco-2 cells cultured with different (0.1527) g/cm p = 0.0012 and 0.1914 (0.0545) g/cm p = 0.0013], concentrations of our hydrogel showed increased proliferation, respectively (Figure 6D). reaching a top at 10% when compared to the positive control of the experiment (normal culture medium) (Figure 3). Hemolysis Chronic EC Animal Study rate caused by our hydrogel on blood from three different rats There was variation in both groups in the degree of colitis was 2.829 ± 1.135%. induction throughout the 4 TNBS applications. Figure 7 There was no sign of inflammation, toxicity, or tissue shows representative endoscopic and macroscopic images of malformation on major organs induced by the hydrogel. both groups. Frontiers in Medicine | www.frontiersin.org 6 August 2020 | Volume 7 | Article 415 Bon et al. New Endoscopic Drug Eluting Platform FIGURE 3 | Cell viability of Caco-2 cells in conditioned medium with 1, 2, 5, 10, and 15% of CoverGel. Cells were cultured at 37◦ C and 5% CO2 for 24, 48, 72, and 96 h. Values are expressed in Relative Fluorescence Units (RFU). No statistically significant differences were observed in in the mucosal restoration but would not be enough to cause ponderal evolution at any day of the study (Figure 8A), in BT to tumor development. liver (two animals out of six in each group, 33.3%) (Figure 8B), Mucosal lesions refractory to biological treatments in IBD and in the histological score for CoverGel + IFX (5.48 ± 0.55) patients is a clinical challenge and unmet need that requires and IFX s.c. (4.96 ± 0.71) (Figure 8C). Colon weight/length ratio new treatment modalities. Drug delivery directly to the colon at day 28 was significantly lower in the IFX subcutaneous treated is an option, thanks to oral dosage formulations that are group compared to the group treated with CoverGel + IFX (0.213 designed to achieve a controlled drug release. However, this (0.124) vs. 0.3497 (0.129), p = 0.0087) (Figure 8D). controlled release has mainly been achieved with small molecules The evolution of levels of ATIs for each group is shown in but not for macromolecules, such as biologics (10). Direct Figure 9. At day of euthanasia, levels of ATIs were significantly release in the colon is beneficial to prevent early absorption lower in rats treated with CoverGel + IFX compared to rats of the drug, thus increasing the quantity of drug reaching treated with IFX subcutaneously (0.90 ± 0.06 vs. 1.97 ± 0.66 the target tissue and at the same time decreasing systemic µg/mL-c, p = 0.0025). exposure and associated side-effects. A specific form of control drug release is polymer-based drug delivery. Polymers offer a DISCUSSION wide range of possibilities useful for colonic delivery. These polymers can be triggered to start releasing drug by physiological In this paper we have reported the development and characteristics of the healthy colon, as well as in different characterization of a new hydrogel, the feasibility of the disease situations (10). Hydrogels for drug delivery have also application of this hydrogel as a drug-eluting platform, and been studied for their capacity of mucoadhesion and controlled the effectiveness of this platform to treat local inflammatory delivery (11). lesions in the colon. This new hydrogel is based on Poloxamer Local injection of infliximab can be a novel scenario for Pluronic-F127, Hyaluronic acid, and Methylcellulose which endoscopic therapy in IBD patients with symptomatic isolated present biocompatible, bioadhesive, and bioactive properties. mucosal lesions, as has been proposed by our group (12). Alone, these three components of which our hydrogel is The real efficacy of locally injected IFX is not discernible composed have proven their biocompatibility among other from current literature, but available data seem encouraging beneficial characteristics (10–14). HA is well-known for its at a dose of 100 mg per session, assuming that higher anti- promotion of cellular viability which is thought to be via the TNF concentrations given through local injection can favor CD44 receptor and the IL6 pathway; this effect has also been healing and prevent complications, because a trend for a dose– seen in tumor cells (15–17). This could represent a problem response curve was observed. Topical injection of anti-TNF since an inflammatory environment such as a UC increase agents has been described in the treatment of long-standing the risk of tumor development. However, despite this, HA perianal fistulizing CD in the absence of an abscess (2). There are hydrogels loaded with antitumoral drugs have been very useful technical variations in topical injection of the anti-TNF agents. in inducing cancer cell death (18) in truly tumor environments, The agents were injected in a circumferential or intrafistula so we assumed that this cellular viability enhancement helps fashion via the external or internal. Frontiers in Medicine | www.frontiersin.org 7 August 2020 | Volume 7 | Article 415 Bon et al. New Endoscopic Drug Eluting Platform FIGURE 4 | In vivo biocompatibility study. (A) % of organ weight relative to total body weight after intraperitoneal injection of 10 mL/kg of CoverGel on rats. represents control rats rats euthanized 3 days post injection and rats euthanized 7 days post injection. (B) Representative images of major organs H/E stainings from each group at a 40 × zoom. (C) H/E staining images of tissue surrounding subcutaneously placed CoverGel after 10 days post placement (10 × left image, 40 × right image). (D) Scanning electron microscopy images of hydrogel after 10 days post subcutaneous placement. In this work, we have provided a new therapeutic tool for and waste flow—essential characteristics for cell survival within endoscopy by developing a new hydrogel (CoverGel) based on the hydrogel. This platform is less vulnerable to matrix HA, MC, and F127, which presents biocompatible, bioadhesive, transformations due to osmotic pressure or tissue extrusion, and bioactive properties (13–17). CoverGel is an endoscopic which is an advantage for a controlled drug release of the vehicle to locally deliver biological drugs (IFX and VDZ) with hydrogel (19). Different kinetic models were studied with proven efficacy in acute and chronic EC in rats. Moreover, the the results obtained from the drug release experiments. The endoscopic administration of IFX induces significantly lower kinetic model that best fitted our results was the Higuchi levels of ATI as compared with subcutaneous administration. The release model. This model is explained by Fick’s law, in which rationale for the drug dosages used (1 mg/mL) with our drug- the diffusion of a molecule is related to its concentration. eluting gel is based on previous clinical experience with local The diffusion flux goes from a higher concentrated area to injection of IFX (12). a lesser concentrated one. The magnitude of the diffusion Our developed hydrogel showed a reverse thermosensitive flux will be proportional to the concentration gradient. These capacity of gelation accompanied by a greater adhesion capacity. results and kinetic model were confirmed when mechanistic of When the hydrogel contacts the mucosal layer at a physiological drug release from Covergel were studied by the Korsmeyer- temperature, it gelifies permitting the adhesion to the mucosa Peppas model, confirming that the diffusion coefficient follows (18). Rheological studies in our hydrogel confirmed that it a Fickian diffusion. Our hydrogel showed the capacity to is a non-Newtonian fluid, meaning that its viscosity varies improve cellular viability in a dose-dependent manner at with temperature. Morphologically, our hydrogel presents a least until a 10%, demonstrating a good potential to become macroporous (pore diameters > 50 nm) interconnected network a cell carrier. Intraperitoneal injection stated that CoverGel that can be the matrix for cell ingrowth as well as nutrients did not cause an acute toxicity reaction on animals 3 or 7 Frontiers in Medicine | www.frontiersin.org 8 August 2020 | Volume 7 | Article 415 Bon et al. New Endoscopic Drug Eluting Platform FIGURE 5 | Endoscopic and macroscopic representative images of the acute EC animal model. Endoscopic images from day 3 (day of treatment) and day 7 (day of euthanasia) and macroscopic image of day 7 (day of euthanasia) from two representative animals of each group. days post intraperitoneal injection. Subcutaneous placement of All of these results help us conclude that this hydrogel is a our hydrogel in rats was done to evaluate in vivo behavior good candidate to be used in the digestive system and implement and biodegradability of our hydrogel, which was evaluated in endoscopic techniques in order to cover some unmet needs by SEM, and we confirmed that the inner structure was of the field. Although in vitro and in vivo results are really maintained 10 days after s.c. placement and no cell infiltration promising, further experiments on in vivo drug release kinetics was seen. and degradation need to be performed, as these characteristics Frontiers in Medicine | www.frontiersin.org 9 August 2020 | Volume 7 | Article 415 Bon et al. New Endoscopic Drug Eluting Platform FIGURE 6 | Clinical evaluations on rats from the acute EC animal model. Color representations are as follows: Control group Sham group CoverGel group CoverGel + IFX group and CoverGel + VDZ group. (A) Variation (1) of weight relative to day 0 (day of induction. (B) Bacterial translocation to the liver. (C) Histological score. (D) Colon weight/length ratio. *p < 0.05; **p < 0.01; ***p < 0.005; ****p < 0.001. have previously been reported to vary from in vitro analysis to in The efficacy of an enema based on ascorbyl palmitate hydrogel vivo, due to inflammatory cell response (20–22). loaded with dexamethasone has been demonstrated in mice TNBS-induced colitis has been proven as a valid method with colitis as well as in the ex vivo analysis of colon samples to elucidate different mechanisms underlying the mechanism from patients with IBD (24). By contrast, our hydrogel, which of the disease (23). Furthermore, and besides its limitations can be also administered as enema, has been designed to be mimicking all different aspects of this complex disease, which administered under an endoscopic view to perform a more are still unknown, this model is a great tool to study the localized treatment. Another goal of our project was to improve immunopathogenesis of the disease and the potential benefit of the patient developing resistance to anti-TNF agents, which different therapies (23). is one of the main problems associated with biologics. This Our results show that the simple application of our hydrogel resistance is caused by the formation of antidrug antibodies as a dressing on the inflamed tissue by means of a shield (ADAs), as a response of the body to these exogenous complex improves the overall outcome and restoration of the mucosal proteins. The kind of mAbs most used in IBD, no matter their layer. This beneficial effect of CoverGel is based on its good nature, cause to a greater or lesser degree ADA formation (25). biocompatibility properties, able to improve mucosal healing. Since new drug development is a complex and long-lasting However, when CoverGel is loaded with biological drugs with effort, new therapeutic approaches are needed to reduce this proven efficacy treating this disease, the effect is even more risk and allow patients to continue on a successful therapy pronounced. In this sense, endoscopic response was obtained longer. The direct drug administration with CoverGel was able after 4 days of treatment in all animals treated with CoverGel, to significantly improve the disease outcome in an animal but unfortunately, histological remission was not achieved model using a much lower dose than the standard dose at this time. Despite that we do not have measurements applied. Moreover, when we compared the formation of ATIs of local and systemic levels of the drugs applied by the from subcutaneous injection vs. CoverGel application in an hydrogel, clinical response was accompanied by different clinical experimental model of chronic EC (with a longer period of aspects such as body weight loss, colon weight/length ratio, time of the disease state), we observed that whereas there were or bacterial translocation to the liver; all features improved no major significant differences on the outcome of the disease in the groups treated with CoverGel alone or in combination between both ways of treatment application, the levels of ATIs with drugs. were significantly lower on the group treated with CoverGel. Frontiers in Medicine | www.frontiersin.org 10 August 2020 | Volume 7 | Article 415 Bon et al. New Endoscopic Drug Eluting Platform FIGURE 7 | Endoscopic and macroscopic representative images of the chronic EC animal model. Endoscopic images from day 3 (first treatment application), day 7 (second TNBS induction), day 21 (fourth TNBS induction), day 24 (fourth treatment application), and macroscopic image of day 28 (day of euthanasia). These results confirmed our theory that a locally applied therapy ADAs are lower with this approach, opening the opportunity at a lower dose is able to obtain a good clinical result and at the to be used in patients with a past history of acute infusion same time reduce adverse events and complications associated reactions to the drug. Further research, preclinical and clinical, with the systemic application of a drug. needs to be pursued to safely evaluate and introduce this Taken together, the results from our study, although technique in standard practice to increase the pool of therapeutic preliminary and preclinical, demonstrate that endoscopic interventions to treat IBD patients. Direct drug administration administration of a based-on hydrogel drug delivery platform through endoscopy opens new possibilities for therapeutic represents a safe and promising approach for the management endoscopy, allowing for the application of bioactive treatments of active lesions with a lower dose as compared with systemic in different pathologies already managed with endoscopy such as administration. Present findings also suggest than levels of colorectal cancer and local inflammatory lesions. Frontiers in Medicine | www.frontiersin.org 11 August 2020 | Volume 7 | Article 415 Bon et al. New Endoscopic Drug Eluting Platform FIGURE 8 | Clinical evaluations on rats from the acute EC animal model. Color representations are as follow: CoverGel + IFX group and IFX s.c. group. (A) Variation (1) of weight relative to day 0 (day of induction). (B) Bacterial translocation to the liver. (C) Histological score. (D) Colon weight/length ratio. **p < 0.01. FIGURE 9 | Evolution of antibodies to Infliximab (ATIs) in serum for: CoverGel + IFX group and IFX s.c. group. Graph represents the concentration of ATIs (µg/mL-c calibrated in comparison to a polyclonal rabbit α-IgG standard). *p < 0.05. Frontiers in Medicine | www.frontiersin.org 12 August 2020 | Volume 7 | Article 415 Bon et al. New Endoscopic Drug Eluting Platform DATA AVAILABILITY STATEMENT FUNDING All datasets generated for this study are included in the This work was supported by project PI16/01928 given to VL-Z article/supplementary material. by the Health Institute Carlos III www.isciii.es and by the XIV GETECCU-Otsuka award given to VL-Z by the Spanish group ETHICS STATEMENT for the study of Crohn and ulcerative colitis www.geteccu.org. ICN2 (MC-S) was supported by the Severo Ochoa program from The animal study was reviewed and approved by Animal Spanish MINECO (Grant No. SEV-2017-0706) www.mineco. Experimentation Ethics Committee of the Hospital Universitari gob.es. Germans Trias I Pujol (registered as B9900005). ACKNOWLEDGMENTS AUTHOR CONTRIBUTIONS The authors would like to acknowledge the help of the personnel IB, RB, and VL-Z conceived and designed the experiments. IB, in the animal facility of the Health Research Institute Germans MC-S, RB, and VL-Z performed the experiments. IB and NO Trias I Pujol and to the Pharmacy Unit and the Biochemistry Unit analyzed the data. RB, VL-Z, and MC-S contributed reagents, in the University Hospital Germas Trias I Pujol. We would also materials, and analysis tools. IB, RB, and VL-Z wrote the like to thank the Endoscopy Unit and the Digestive Service at the paper. All authors contributed to the article and approved the University Hospital Germans Trias I Pujol. This manuscript has submitted version. been released as a pre-print at Biorxiv, Bon et al. (26). REFERENCES 12. Caicco MJ, Zahir T, Mothe AJ, Ballios BG, Kihm AJ, Tator CH, et al. Characterization of hyaluronan-methylcellulose hydrogels for cell delivery 1. Paramsothy S, Rosenstein AK, Mehandru S, Colombel JF. The current to the injured spinal cord. J Biomed Mater Res A. (2013) 101:1472– state of the art for biological therapies and new small molecules 7. doi: 10.1002/jbm.a.34454 in inflammatory bowel disease. Mucosal Immunol. (2018) 11:1558– 13. Law N, Doney B, Glover H, Qin Y, Aman ZM, Sercombe 70. doi: 10.1038/s41385-018-0050-3 TB, et al. Characterisation of hyaluronic acid methylcellulose 2. Shen B, Kochhar G, Navaneethan U, Liu X, Farraye FA, Gonzalez-Lama hydrogels for 3D bioprinting. J Mech Behav Biomed Mater. (2018) Y, et al. Role of interventional inflammatory bowel disease in the era 77:389–99. doi: 10.1016/j.jmbbm.2017.09.031 of biologic therapy: a position statement from the Global Interventional 14. Diniz IMA, Chen C, Xu X, Ansari S, Zadeh HH, Marques MM, et al. IBD Group. Gastrointest Endosc. (2019) 89:215–37. doi: 10.1016/j.gie.2018. Pluronic F-127 hydrogel as a promising scaffold for encapsulation of 09.045 dental-derived mesenchymal stem cells. J Mater Sci Mater Med. (2015) 3. Bon I, Bartolí R, Lorenzo-Zúñiga V. Endoscopic shielding technique, 26:153. doi: 10.1007/s10856-015-5493-4 a new method in therapeutic endoscopy. World J Gastroenterol. (2017) 15. Vincent T, Jourdan M, Sy MS, Klein B, Mechti N. Hyaluronic acid induces 23:3761. doi: 10.3748/wjg.v23.i21.3761 survival and proliferation of human myeloma cells through an interleukin- 4. Lorenzo-Zúñiga V, Boix J, Moreno de Vega V, Bon I, Marín I, Bartolí R. 6-mediated pathway involving the phosphorylation of retinoblastoma Endoscopic shielding technique with a newly developed hydrogel to prevent protein. J Biol Chem. (2001) 276:14728–36. doi: 10.1074/jbc.M0039 thermal injury in two experimental models. Dig Endosc. (2017) 29:702– 65200 11. doi: 10.1111/den.12864 16. Matsui Y, Inomata M, Izumi K, Sonoda K, Shiraishi N, Kitano S. Hyaluronic 5. Hoffman AS. Hydrogels for biomedical applications. Adv Drug Deliv Rev. acid stimulates tumor-cell proliferation at wound sites. Gastrointest Endosc. (2002) 54:3–12. doi: 10.1016/S0169-409X(01)00239-3 (2004) 60:539–43. doi: 10.1016/S0016-5107(04)01890-5 6. Buwalda SJ, Boere KWM, Dijkstra PJ, Feijen J, Vermonden 17. Murakami T, Otsuki S, Okamoto Y, Nakagawa K, Wakama H, Okuno N, et al. T, Hennink WE. Hydrogels in a historical perspective: from Hyaluronic acid promotes proliferation and migration of human meniscus simple networks to smart materials. J Control Release. (2014) cells via a CD44-dependent mechanism. Connect Tissue Res. (2018) 60:17– 190:254–73. doi: 10.1016/j.jconrel.2014.03.052 27. doi: 10.1080/03008207.2018.1465053 7. Ahmed EM. Hydrogel : preparation, characterization, and applications : a 18. Moraima Morales-Cruz CMF, Jean C Fernández BNS. Induction of cancer review. J Adv Res. (2015) 6:105–21. doi: 10.1016/j.jare.2013.07.006 cell death by hyaluronic acid-mediated uptake of cytochrome C. J Nanomed 8. Varaprasad K, Malegowd G, Jayaramudu T, Mohan M, Sadiku R. A Nanotechnol. (2015) 6:316. doi: 10.4172/2157-7439.1000316 mini review on hydrogels classi fi cation and recent developments in 19. Jiang G, Sun J, Ding F. PEG-g-chitosan thermosensitive hydrogel for miscellaneous applications. Mater Sci Eng C Mater Biol Appl. (2017) 79:958– implant drug delivery: cytotoxicity, in vivo degradation and drug release. 71. doi: 10.1016/j.msec.2017.05.096 J Biomater Sci Polym Ed. (2014) 25:241–56. doi: 10.1080/09205063.2013.8 9. Erben U, Loddenkemper C, Doerfel K, Spieckermann S, Haller D, 51542 Heimesaat MM, et al. A guide to histomorphological evaluation of intestinal 20. Molinaro G, Leroux JC, Damas J, Adam A. Biocompatibility of inflammation in mouse models. Int J Clin Exp Pathol. (2014) 7:4557–76. thermosensitive chitosan-based hydrogels: an in vivo experimental 10. Tam RY, Cooke MJ, Shoichet MS. A covalently modified hydrogel approach to injectable biomaterials. Biomaterials. (2002) 23:2717– blend of hyaluronan-methyl cellulose with peptides and growth factors 22. doi: 10.1016/S0142-9612(02)00004-2 influences neural stem/progenitor cell fate. J Mater Chem. (2012) 22:19402– 21. Jin R, Moreira Teixeira LS, Dijkstra PJ, Karperien M, van 11. doi: 10.1039/c2jm33680d Blitterswijk CA, Zhong ZY, et al. Injectable chitosan-based 11. Gupta D, Tator CH, Shoichet MS. Fast-gelling injectable blend of hyaluronan hydrogels for cartilage tissue engineering. Biomaterials. (2009) and methylcellulose for intrathecal, localized delivery to the injured 30:2544–51. doi: 10.1016/j.biomaterials.2009.01.020 spinal cord. Biomaterials. (2006) 27:2370–9. doi: 10.1016/j.biomaterials.2005. 22. Sun J, Jiang G, Wang Y, Ding F. Thermosensitive chitosan hydrogel 11.015 for implantable drug delivery: blending PVA to mitigate body Frontiers in Medicine | www.frontiersin.org 13 August 2020 | Volume 7 | Article 415 Bon et al. New Endoscopic Drug Eluting Platform response and promote bioavailability. J Appl Polym Sci. (2012) Conflict of Interest: VL-Z and RB are coauthors of a patent for the 125:2092–101. doi: 10.1002/app.36297 “BIOADHESIVE PLATFORM TO PERFORM BIOACTIVE TREATMENT” 23. Antoniou E, Margonis A, Angelou A, Pikouli A, Argiri P, Karavokyros I, et al. International application number “PCT/EP2017/068876” and International The TNBS-induced colitis animal model: an overview. Ann Med Surg. (2016) Publication Number “WO 2018/019881 A1” placed in the WIPO/PCT (World 11:9–15. doi: 10.1016/j.amsu.2016.07.019 Intellectual Property Organization/Patent Cooperation Treaty). 24. Zhang S, Ermann J, Succi MD, Zhou A, Hamilton MJ, Cao B, et al. An inflammation-targeting hydrogel for local drug The remaining authors declare that the research was conducted in the absence of delivery in inflammatory bowel disease. Sci Transl Med. (2015) any commercial or financial relationships that could be construed as a potential 7:300ra128. doi: 10.1126/scitranslmed.aaa5657 conflict of interest. 25. Vermeire S, Gils A, Accossato P, Lula S, Marren A. Immunogenicity of biologics in inflammatory bowel disease. Therap Adv Copyright © 2020 Bon, Cano-Sarabia, de la Ossa, Bartolí and Lorenzo-Zúñiga. Gastroenterol. (2018) 11:1756283X1775035. doi: 10.1177/1756283X177 This is an open-access article distributed under the terms of the Creative Commons 50355 Attribution License (CC BY). The use, distribution or reproduction in other forums 26. Bon I, Cano-Sarabia M, de la Ossa, de N, Bartolí R, Lorenzo-Zúñiga V. is permitted, provided the original author(s) and the copyright owner(s) are credited Development and characterization of a new endoscopic drug eluting platform and that the original publication in this journal is cited, in accordance with accepted with proven efficacy in acute and chronic experimental colitis. BioRxiv. academic practice. No use, distribution or reproduction is permitted which does not (2019). doi: 10.1101/712224. [Epub ahead of print]. comply with these terms. Frontiers in Medicine | www.frontiersin.org 14 August 2020 | Volume 7 | Article 415