Development of a novel injectable drug delivery system for subconjunctival glaucoma treatment

    Development of a novel injectable drug delivery system for subconjunctival glaucoma treatment Karsten Voss, Karen Falke, Arne Bernsdorf, Niels Grabow, Christian Kastner, Katrin Sternberg, Ingo Minrath, Thomas Eickner, Andreas Wree, Klaus-Peter Schmitz, Rudolf Friedrich Guthoff, Martin Witt, Marina Hovakimyan PII: DOI: Reference:

S0168-3659(15)30008-0 doi: 10.1016/j.jconrel.2015.06.035 COREL 7740

To appear in:

Journal of Controlled Release

Received date: Revised date: Accepted date:

25 March 2015 26 June 2015 27 June 2015

Please cite this article as: Karsten Voss, Karen Falke, Arne Bernsdorf, Niels Grabow, Christian Kastner, Katrin Sternberg, Ingo Minrath, Thomas Eickner, Andreas Wree, Klaus-Peter Schmitz, Rudolf Friedrich Guthoff, Martin Witt, Marina Hovakimyan, Development of a novel injectable drug delivery system for subconjunctival glaucoma treatment, Journal of Controlled Release (2015), doi: 10.1016/j.jconrel.2015.06.035

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT 1 Development of a novel injectable drug delivery system for subconjunctival glaucoma treatment Karsten Vossa#, Karen Falkeb#, Arne Bernsdorfa, Niels Grabowa, Christian Kastnera, Katrin

a

IP

SC R

Friedrich Guthoffa, Martin Wittc, Marina Hovakimyana

T

Sternberga, Ingo Minratha, Thomas Eicknera, Andreas Wreec, Klaus-Peter Schmitza, Rudolf

Institute for Biomedical Engineering, Rostock University Medical Center, Friederich-

Barnewitz-Strasse 4, D-18119 Rostock, Germany b

Department of Ophthalmology, Rostock University Medical Center, Doberaner Strasse 140,

NU

D-18057 Rostock, Germany c

Department of Anatomy, Rostock University Medical Center, Gertrudenstrasse 9a, D-18057

MA

Rostock, Germany #-equal contributors

D

[email protected], [email protected], [email protected],

TE

[email protected], [email protected], [email protected], [email protected],

[email protected], [email protected],

CE P

[email protected], [email protected], [email protected]

AC

Corresponding author:

Marina Hovakimyan, PhD Institute for Biomedical Engineering Rostock University Medical Center University of Rostock Friederich-Barnewitz-Strasse 4, D-18119 Rostock Germany

Phone: +49-381-54345547 Fax: +49-381-54345602 Email: [email protected]

Running headline: Drug delivery system for glaucoma treatment

ACCEPTED MANUSCRIPT 2 Abstract In this study we present the development of an injectable polymeric drug delivery system for subconjunctival treatment of primary open angle glaucoma. The system consists of hyaluronic acid sodium salt (HA), which is commonly used in ophthalmology in anterior segment surgery,

T

and an isocyanate - functionalized 1,2-ethylene glycol bis(dilactic acid) (ELA-NCO). The polymer

IP

mixtures with different ratios of HA to ELA-NCO (1/1, 1/4, and 1/10 (v/v)) were investigated for

SC R

biocompatibility, degradation behavior and applicability as a sustained release system. For the latter, the lipophilic latanoprost ester pro-drug (LA) was incorporated into the HA/ELA-NCO system.

In vitro, a sustained LA release over a period of about 60 days was achieved. In cell culture

NU

experiments, the HA/ELA-NCO (1/1, (v/v)) system was proven to be biocompatible for human and rabbit Tenon’s fibroblasts. Examination of in vitro degradation behavior revealed a total

MA

mass loss of more than 60% during the observation period of 26 weeks. In vivo, LA was continuously released for 152 days into rabbit aqueous humor and serum. Histological investigations revealed a marked leuko-lymphocytic infiltration soon after

D

subconjunctival injection. Thereafter, the initial tissue reaction declined concomitantly with a continuous degradation of the polymer, which was completed after 10 months.

TE

Our study demonstrates the suitability of the polymer resulting from the reaction of HA with ELA-NCO as an injectable local drug delivery system for glaucoma therapy, combining

CE P

biocompatibility and biodegradability with prolonged drug release.

Keywords: glaucoma, local drug delivery, injectable polymer, hyaluronic acid, 1,2-ethylene

AC

glycol bis(dilactic acid)

ACCEPTED MANUSCRIPT 3 1. Introduction Glaucoma encompasses a heterogenic group of ophthalmic diseases that damage the optic nerve, resulting in gradual visual impairment and potentially irreversible vision loss. Worldwide, over 60 million individuals are affected, and this number is expected to rise to 80

T

million by 2020 [1]. Elevated intraocular pressure (IOP) is often associated with primary open

IP

angle glaucoma. Results from long-term clinical studies show that lowering the IOP to a normal

SC R

level can slow down or even stop the progression of the disease [2]. This can be achieved either surgically or pharmacologically with eye drops, which reduce the production of the aqueous humor (e.g. carbonic anhydrase inhibitors and β-adrenergic receptor antagonists) and/or increase

its

outflow

(e.g.

α2-adrenergic

receptor

agonists,

parasympathomimetics,

NU

sympathomimetics, and prostaglandin analogues) [3]. Topically instilled drugs penetrate through the corneal or the scleral route, additionally some conjunctival contribution is observed [4].

MA

A major disadvantage of eye drops is poor patient adherence to daily medication dosing instructions. Especially elderly people have difficulties with routine daily applications which are necessary to reach a permanent IOP decrease. Approximately 20% of glaucoma patients exhibit

D

poor treatment compliance [5,6]. Previous studies have demonstrated that inadequate patient adherence is often associated with more severe visual field loss [7]. The common use of

TE

preservatives like benzalkonium chloride in the antiglaucoma medication has been shown to exert cytotoxic and inflammatory effects on corneal and conjunctival tissue [8]. Studies have

CE P

shown that long-term application of glaucoma eye drops may result in ocular discomfort, dry eye, burning, foreign body sensation, and red eye [9]. Moreover, medications used to treat glaucoma can be absorbed systematically and induce clinically relevant systemic effects [10].

AC

A promising approach addressing poor compliance and resulting fluctuations of IOP is the administration of drugs via local drug delivery (LDD) systems, thereby ensuring a sustained drug concentration over an extended period of time. Continuous drug release will reduce the need for daily drug administration which could improve patient adherence and treatment outcome. Using such systems could also be more economic than application of eye drops since smaller amounts of drugs might be needed to achieve the same effect. In medical literature, different approaches have been described utilizing topical, implantable, and injectable LDD systems [11]. Topical ocular systems administer drugs from the outside of the eye, while implantable and injectable systems release drugs within the eye. Examples for topical applications are mucoadhesive formulations, hydrogels and particles [12,13]. Latanoprost-loaded contact lenses have been shown to release the drug over a period of up to 4 weeks [14]. Implantable and injectable devices can extend the release time up to several weeks or even months [15]. Subconjunctival administration is well tolerated in patients, considered to be safe, and is already used routinely in clinics for the delivery of different medications. Experimentally, a sustained

ACCEPTED MANUSCRIPT 4 in vivo drug release for up to 4 weeks was shown following subconjunctival implantation of dorzolamide-loaded polymer disks in rabbit eyes [16]. A therapeutic lowering of IOP beyond 50 days has been demonstrated in rabbit experiments after a single subconjunctival injection of latanoprost (LA)-loaded liposomes [17].

T

An ideal LDD is expected to combine consistent delivery of an appropriate drug dosage for up to

IP

4 months with a complete degradation within this time and minimal side effects on the

SC R

surrounding tissue.

Within this context, the objective of the present work was to develop a drug-loaded, biodegradable, and biocompatible LDD system which would provide localized, long-term release of IOP-lowering medication. The developed LDD system consists of 2 components: hyaluronic

NU

acid sodium salt (HA) and a more hydrophobic hexamethylene diisocyanate (HDI)-functionalized 1,2-ethylene glycol bis(dilactic acid) (ELA-NCO). HA is a naturally occurring polysaccharide with

MA

distinct physicochemical characteristics which is commonly used in ophthalmic microsurgery due to its viscoelastic and hydrophilic properties [18]. ELA-NCO has been previously introduced by our group as a biodegradable tissue adhesive with high adhesive strength and good in vivo

D

biocompatibility [19,20].

Hydrophobic latanoprost ester pro-drug (LA), which is successfully used in glaucoma therapy,

TE

was selected as IOP-lowering model drug to be incorporated into the HA/ELA-NCO system. Here, we describe the drug release of LA-loaded HA/ELA-NCO polymer samples as well as the

CE P

degradation behavior in vitro and in vivo. The biocompatibility of the LDD system was investigated in vitro using ocular fibroblasts and in vivo after subconjunctival injection in rabbits.

AC

2. Material and methods 2.1 Materials

Hyaluronic acid sodium salt (HA) from Streptococcus equi was purchased from SigmaAldrich (Taufkirchen, Germany, Cat. # 53747). 1,2-ethylene glycol bis(dilactic acid) (ELA) was prepared as described by Heiss and colleagues [21]. In brief, to a mixture of 1 mol of ethylene glycol and 2 mol of lactide 1.5 g of 85% phosphoric acid was added as catalyst. After heating at 100 °C for 1 hour the temperature was raised to 130 °C and held for at least 6 hours. Subsequently, ELA was endowed with terminal reactive isocyanate groups by a stoichiometric reaction (ratio of 1/2 (n/n)) with aliphatic hexamethylene diisocyanate (HDI) at 50 °C to yield ELA-NCO according to Sternberg and colleagues [22]. To reduce its viscosity and enhance handling, ELA-NCO was diluted with absolute dimethyl sulfoxide (DMSO, 15% w). Poly(L-lactide) (PLLA, Resomer® L214, Mw = 650,000 g/mol) was purchased from Boehringer Ingelheim Pharma GmbH & Co. KG (Ingelheim, Germany).

ACCEPTED MANUSCRIPT 5 Lysozyme from chicken egg white (45100–48200 U/mg), organic solvents, and all other reagents were purchased from Sigma-Aldrich (Taufkirchen, Germany). Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 4.5 g/L glucose was used from AppliChem (Darmstadt, Germany), whereas fetal calf serum, penicillin G, and

Fourier-Transform-Infrared (FTIR) spectroscopy

SC R

2.2

IP

T

streptomycin were obtained from PAA Laboratories GmbH (Cölbe, Germany).

To monitor the polymerization reaction between HA and ELA-NCO FTIR spectroscopy (spectral range between 4,000 and 700 cm-1, EQUINOX 55 FTIR equipped with ATR measuring cell, Bruker Optik GmbH, Ettlingen, Germany) was used. The samples were

NU

stored in sealable glass vials at 20 °C before measurement. The in situ reaction progress

2.3

MA

was traced by decrease in the isocyanate’s (–NCO group) absorption band at 2,270 cm-1.

Drug incorporation and in vitro release

For drug incorporation, LA was prepared as suspension of 250 mg LA in 9.5 g water before

D

250 mg HA were added while stirring to constitute HA(LA). For drug release studies, the concentration of LA in eluates from HA(LA)/ELA-NCO samples,

TE

with ratios of 1/1, 1/4 and 1/10 (v/v), was measured at defined time points. In vitro drug release was investigated at 37 °C. Samples of HA(LA)/ELA-NCO (n = 6) were

CE P

spread inside glass vials with double chamber syringes attached to a mixing extruder (Sulzer Mixpac AG, Rotkreuz, Switzerland), and covered immediately with 2 mL eluent (0.9% NaCl solution). The eluent was changed daily, and theLA concentration in the eluent was

AC

determined by HPLC (Eurospher column 100-5, C18, 120 x 4 mm ID, Wissenschaftlicher Gerätebau Dr.-Ing. Herbert Knauer GmbH, Berlin, Germany). 2.4

In vitro degradation behavior

The in vitro degradation behavior of the polymeric drug carrier HA/ELA-NCO was investigated for a period of 26 weeks. As reference material PLLA was used, since ELA-NCO also contains dilactide units. The degradation study was performed at 37 °C in Sørensen buffer solution (0.1 M, pH 7.4) supplemented with lysozyme (1.5 µg/mL). To prevent microbiological growth, penicillin G (100 U/mL) and streptomycin (100 µg/mL) were added. For sample preparation HA/ELA-NCO with mixing ratios of 1/1, 1/4 and 1/10 (v/v) were applied with double chamber syringes to Teflon molds (diameter = 25 mm, depth = 500 µm) to obtain films with reproducible surface/volume ratios. PLLA films were manufactured in solution-based casting processes from chloroform. The sample films were first dried in vacuo at 40 °C for 2 days to eliminate any residual solvents, and then cut with a cork trepan in order to obtain samples with a defined diameter of 6 mm and a thickness of 500 µm. All samples

ACCEPTED MANUSCRIPT 6 were sterilized with ethylene oxide prior to use. To reduce the effect of EO adsorption, the sterilized samples were extensively degassed in vacuo before being stored in the degradation solution, which was changed daily during the course of the experiment. For each material and degradation period n = 6 samples were analyzed. The chosen

T

degradation intervals were 1, 2, 4, 6, 8, 12, 16, 20, and 26 weeks. The polymer samples

IP

were agitated (70 rpm) at 37 °C and the medium was changed daily due to lysozyme

SC R

instability [23]. To quantify the degradation behavior, the dry mass of each sample was determined before and after storage in the degradation medium. The samples were washed with demineralized water, dried at room temperature in a desiccator for 48 h at 30 mbar and

2.5

NU

weighed against a constant weight (n = 3 replicate measurements for each sample).

Micro-computer tomography (µ-CT)

MA

For determination of degradation-induced changes in the porosity of the polymer system HA/ELA-NCO, dried samples were subjected to micro-computer tomography (µ-CT) at baseline and after 26 weeks storage in the degradation medium, as described above. The

D

investigations were performed by means of high-resolution µ-CT (SkyScan 1172, SkyScan N.V., Aartselaar, Belgium) equipped with a 80 kV microfocus X-ray source (Hamamatsu

TE

Photonics K.K., Iwata-City Shizuoka, Japan). Following µ-CT conditions were chosen: high voltage (80 kV), electron current (100 µA), pixel size (11.37 µm), rotation range (180°) and

CE P

rotation step (0.9°). No filter was used. The total porosity, defined as volume of all open and closed pores and specified as percent of the total volume-of-interest of the samples, was

2.6

AC

determined by the software CT Analyser (version 1.10.0.2, SkyScan N.V.) Cell culture

Human Tenon’s capsule fibroblasts tissue cultures were established using specimens obtained during strabismus surgery (Department of Ophthalmology, University of Rostock, Germany) as described elsewhere [24]. The study protocol was approved by the institutional Ethics Committee. The tenets of the Declaration of Helsinki were followed and written informed consent was obtained from all participants. Rabbit Tenon’s fibroblasts were obtained from corneoscleral rims of New Zealand White rabbits (Charles River GmbH, Sulzfeld, Germany). Human and rabbit Tenon’s fibroblasts were used between passages 3 to 5 and cultured in 75 cm² flasks under standard conditions (37 °C, 5% CO2 and relative humidity of 95%) in a cell incubator. Prior to use, DMEM was supplemented with 4.5 g/L glucose, 2.2 g/L NaHCO3, 10% fetal calf serum, 100 U/mL penicillin G, 100 μg/mL streptomycin, and adjusted to pH 7.4. For eluate preparation pure ELA-NCO samples and freshly mixed HA/ELA-NCO samples (1/1, 1/4, and 1/10 (v/v)) were incubated with serum-free DMEM (0.2 g sample/mL)

ACCEPTED MANUSCRIPT 7 and eluted for 24 h at 37 °C. After elution, the aqueous supernatant was collected, sterile filtered (0.22 µm, Rotilabo(R)-syringe filters, Carl Roth GmbH & Co. KG, Karlsruhe, Germany), and 10% fetal calf serum was added. This mixture is referred to as 100% (v) eluate. The eluate was further diluted with DMEM to concentrations of 50, 25, 12.5, 6.25, and

In vitro biocompatibility

SC R

2.7

IP

T

3.125% (v) according to the demands of DIN EN ISO 10993-5/10993-12.

Cell viability was analyzed by the fluorescent assay CellQuanti-Blue™ (BioAssay Systems, Hayward, USA). Briefly, 24 h after seeding the fibroblasts (2,000 cells/200 µL DMEM) into 96-well microtiter plates, the cells were incubated with the serial dilutions of the sample

NU

eluates for 48 h. For each dilution, measurements were run in quadruplicates. As negative control (NC) cells were incubated with DMEM supplemented with 10% fetal calf serum. After

MA

48 h samples and DMEM were removed and replaced by a 1:10 dilution of the CellQuantiBlue™ reagent in DMEM for 2 h. Thereafter, the fluorescence was measured at 590 nm (excitation at 544 nm). The mean fluorescence intensity value corresponding to the NC was

D

defined as 100%. The relative viability was calculated as percentage of the mean

2.8

TE

fluorescence intensity of treated cells relative to that of the NC. In vivo experiments

CE P

All animal experiments were performed using female, normotensive New Zealand White rabbits in compliance with guidelines of the European Community for the use of experimental animals.

AC

The study protocols were approved by the local Animal Ethics Committee (Approval ID: 7721.3-1.1-017/13).

For in vivo analysis of degradation and biocompatibility 12 rabbits received subconjunctival injections of approximately 50 mg HA/ELA-NCO 1/1 (v/v) using a 27 gauge extruder system, under general anesthesia. One day, one week, and every following month after injection (for up to 10 months) photographic documentation and analysis of histological sections were performed. We adopted the Moorfields Bleb Grading System (MBGS) [25] to characterize pathologic changes in the conjunctiva overlying the injection site. This grading system was developed for the classification of filtering blebs after trabeculectomy, and evaluates morphologic and vascularity parameters based on photographs. In our study we applied the Vascularity Criteria range from 1 to 10 in the center and the periphery (> 2mm from depot edge) to evaluate biomicroscopic images. For a preliminary examination of in vivo drug release, subconjunctival injection of LAincorporated HA(LA)/ELA-NCO 1/1 (v/v) was performed. Up to 100 µl of aqueous humor was

ACCEPTED MANUSCRIPT 8 aspirated from the injected eye via a 29 gauge needle and blood serum was collected at defined time points for the longitudinal measurements of LA concentration for up to 152 days using HPLC-MS.

High performance liquid chromatography-mass spectrometry (HPLC-MS)

T

2.9

IP

Samples (aqueous humor and blood serum) were stored at -20°C and thawed briefly before

SC R

preparation. For protein precipitation 50 µL of the aqueous humor or blood serum were mixed with 50 µL methanol (LCMS grade, Carl Roth GmbH & Co KG, Karlsruhe, Germany), vortexed for 20 s, and stored on ice for 30 min. The mixtures were centrifuged at 12,000 rpm for 20 min at 4°C in a Heraeus Biofuge primo R (Kendro Laboratory Products GmbH,

NU

Langenselbold, Germany). The supernatant was removed and dried under vacuum (Alpha 12/LD plus coupled with RVC-2-33 IR; both Christ, Osterode, Germany). The resulting pellets

MA

were extracted twice with 15 µL methanol. After vortexing for 30 s the samples were centrifuged with 12,000 rpm for 20 min at 4°C. The combined supernatants were analyzed by HPLC-MS. Recovery experiments were conducted and the efficiency of extraction and

D

subsequent detection of LA was 55% in serum and 100% in aqueous humor. Calibration was performed with LA (Cayman Chemicals, Ann Arbor, MI, USA) in methanol

TE

solution diluted to concentrations of 200 ng/mL; 100 ng/mL; 50 ng/mL; 25 ng/mL; 12.5 ng/mL; 6.25 ng/mL; 3.13 ng/mL and 1.56 ng/mL, respectively.

CE P

The HPLC-MS measurements were performed on a Shimadzu LCMS-8030 plus triple quadrupole mass spectrometer (Shimadzu Deutschland GmbH, Duisburg, Germany) equipped with a Nexera UHPLC consisting of 2 liquid delivery modules LC-30AD, a

AC

Prominence DGU-20 A5 online degasser, a SILAC-30AC UHPLC autosampler with a 100 µL-mixing chamber, a thermostated column compartment Prominence CTO-20AC, and a Prominence CBM-20A Interface. A Phenomenex Kinetex 2.6 µm C18 100A, 150x2.1 mm column equipped with a Phenomenex SecurityGuardTM Ultra Guard Cartridge C18, 2.1 mm ID at 40°C was used for chromatographic separation. The mobile phases were 0.1% (v) formic acid in water (A) and 0.1% (v) formic acid in acetonitrile (B). The following gradient was applied at a flow rate of 0.3 mL/min: 0-0.36 min 30% to 70% B, 0.36-2.5 min 70% B, 2.55.5 min 30% B. One microliter of each sample was injected. The ionization of the analyte occurred in an electrospray ionization source in positive mode. Nitrogen was used as drying and nebulizing gas at flow rates of 1.2 L/min and 6 L/min, respectively. The desolvation line temperature was set to 200°C and the heat block temperature was set to 500°C. For quantification of LA the MRM (multiple reaction monitoring) transition m/z 433.3>397.2 was recorded, exhibiting a lower limit of quantification at a latanoprost concentration of 5 ng/mL.

ACCEPTED MANUSCRIPT 9 2.10

Histology

After final biomacroscopical examination, animals were sacrificed with an intravenous overdose of pentobarbital (Sigma-Aldrich Chemie, Munich, Germany). Following enucleation, the globes were fixed overnight in 3.7 % formaldehyde at room temperature. After

T

dehydration in graded ethanols, samples were embedded in paraplast and serially cut into

IP

sections of 6 µm thickness. For general histological observation, the sections were routinely

SC R

stained with hematoxylin-eosin (H&E).

Histological signs of cellular inflammation, number of eosinophils, and giant cells, as well as the presence of a fibrous capsule were scored on a four-point scale according to a modified Wilfingseder grading system [26] (Table 1). Cellularity was divided into low (greater amount

NU

of area accounted for by a few inflammatory cells), moderate (intermediate between low and

MA

high), and high (greater amount of area accounted for by many inflammatory cells) [27].

Phenotype

I

Thin capsule

II

Low cellularity: lymphocytes, eosinophils, absence of giant cells

III

Intermediate cellularity: lymphocytes, eosinophils foreign body giant cells

IV

High cellularity: lymphocytes, eosinophils, foreign body giant cells; neovascularization

Results 3.1

AC

Table 1

CE P

TE

D

Grade

Formation of the polymeric drug carrier

ELA-NCO is equipped with terminal isocyanate groups, which react in a first step with the hydroxyl, amino, and carboxylic acid groups of HA and water to form urethane-groups (Fig. 1A), urea units (Fig. 1B), substituted carbamoyl groups (Fig. 1C), and in the case of water new amino groups (not shown). Further steps may follow in which acidic protons of these formed groups can undergo additional reactions with terminal isocyanate groups (Fig. 1D). Therefore, the concentration of the -NCO groups can be used to monitor the process of the polymerization reaction. This can be done using FTIR spectroscopy measuring the intensity of the -NCO vibration band at ~2,270 cm-1.

ACCEPTED MANUSCRIPT

MA

NU

SC R

IP

T

10

D

Figure 1

TE

At room temperature the forming of the polymer is completed after 48 h for all HA/ELA-NCO mixtures (not shown). Figure 2 depicts the FTIR-ATR spectra of HA/ELA-NCO (1/1 (v/v)) at

AC

CE P

room temperature after reaction times of 2 h, 4 h, 7 h, 24 h, and 48 h (Fig. 2).

Figure 2 3.2

In vitro drug release

For all investigated HA/ELA-NCO mixtures (1/1, 1/4, and 1/10 (v/v)) a sustained LA release over more than 60 days has been achieved (Fig. 3). For the 1/1 (v/v) mixture an initial burst release was observed (Fig. 3, green plot). The mixtures with a higher proportion of the

ACCEPTED MANUSCRIPT 11 hydrophobic component ELA-NCO (blue and red plot for 1/4 and 1/10, respectively) exhibited

MA

NU

SC R

IP

T

a much lower release of LA (Fig. 3).

Figure 3

In vitro degradation, mass loss

D

3.3

TE

To quantify the degradation process the sample mass was determined repeatedly during the 26 weeks of degradation (Fig. 4). For all HA/ELA-NCO (1/1, 1/4, and 1/10 (v/v)) samples a total mass loss of about 65% was observed after 26 weeks. After the first week samples with

CE P

the highest ELA-NCO content (red plot) had retained less than 60% of the initial mass, whereas the mass of samples with the lowest ELA-NCO content still was more than 80% of the initial mass (1/1, green plot). After the first week an almost linear constant mass loss was

AC

observed for all HA/ELA-NCO samples. Overall, the ongoing degradation was higher for the 1/1 (v/v) mixture. For the reference material PLLA, no mass loss was determined within the investigated time frame of 26 weeks (Fig. 4, purple plot).

ACCEPTED MANUSCRIPT

MA

NU

SC R

IP

T

12

Figure 4

In vitro degradation, morphological changes

D

3.4

Morphological changes caused by the degradation procedure were documented macro-

TE

photographically (Fig. 5A, B) and using micro-computer tomography (Fig. 5C). For examination of in vitro degradation behavior, all HA/ELA-NCO samples were initially

CE P

prepared with the same thickness (500 µm) and diameter (6 mm). After 26 weeks of incubation in the degradation medium a decrease in diameter was observed for all sample mixtures (Fig. 5A, right column) compared to the initial morphology

AC

(Fig. 5A, left column). A decrease was also observed in the thickness of sample after 26 weeks of degradation (Fig. 5B, right side) compared to the baseline morphology (Fig. 5B, left side), which was slightly less pronounced for the 1/10 HA/ELA-NCO mixture (Fig. 5B, bottom row). Micro-CT imaging (Fig. 5C) revealed for all HA/ELA-NCO mixtures a total porosity (volume of all open and closed pores, specified as percent of the total volume) of 69% at baseline. After 26 weeks, the samples exhibited a minimal increase in porosity, which was not significant. A porosity increase from 69% to 74% was seen for 1/1 (Fig. 5C, top row) and 1/10 (Fig. 5C, bottom row) HA/ELA-NCO samples, whereas the 1/4 HA/ELA-NCO exhibited an increase from 69 to 70% (Fig. 5C, middle row).

ACCEPTED MANUSCRIPT

CE P

TE

D

MA

NU

SC R

IP

T

13

Figure 5

In vitro biocompatibility of HA/ELA-NCO

AC

3.5

The cell viability assay showed that after 48 h of incubation with undiluted ELA-NCO eluate (100%) the viability of human Tenon’s fibroblasts dropped to 58 ± 16% (Fig. 6A). For all tested dilutions of the ELA-NCO eluate (50, 25, 12.5, 6.25, 3.125% v) viability remained above 75%. The incubation of human Tenon’s finbroblasts the same cells with 100% eluate of 1/10 and 1/4 (v/v) HA/ELA-NCO resulted in a relative cell viability of 43 ± 7% and 63 ± 11%, respectively (Fig. 6B and 6C). None of the dilutions of 1/10 (Fig. 6B) and 1/4 (Fig. 6C) HA/ELA-NCO mixtures exerted cytotoxic effects on human Tenon’s fibroblasts. In case of the 1/1 (v/v) HA/ELA-NCO the cells exhibited only a negligible viability drop (83 ± 6%) after 48 h of incubation with the undiluted HA/ELA-NCO sample (Fig. 6D). The viability of rabbit Tenon’s fibroblasts decreased to about 61 ± 21% after incubation with undiluted eluate (100%) of ELA-NCO (Fig. 7A). Compared to the human Tenon’s cells, the rabbit fibroblasts were more attenuated, when incubated with the undiluted eluates of 1/10 and 1/4 HA/ELA-NCO mixtures; the viability of cells dropped to 16 ± 11% and 18 ± 6%,

ACCEPTED MANUSCRIPT 14 respectively (Fig. 7B and C). However, different dilutions of both, 1/10 and 1/4 HA/ELA-NCO, showed no cytotoxic effects (Fig. 7B and C). As for human Tenon’s fibroblasts, the incubation of the rabbit Tenon’s fibroblasts with different dilutions of 1/1 HA/ELA-NCO mixture revealed no negative influence on cell viability (Fig. 7D). Likewise, a negligible

MA

NU

SC R

IP

T

decrease of viability (86 ± 7%) was observed in the case of the undiluted eluate (Fig. 7D).

Figure 7

F

In vivo drug release

TE

3.6

D

Figure 6

The release profile of LA in aqueous humor (Fig. 8A) revealed an initial burst release within 24 hours of injection with a concentration of 179 ng/mL. On the fifth day after injection, the

CE P

LA concentration dropped to 14 ng/mL, and then continued to decrease, reaching about 8 ng/mL after 152 days post injection (Fig. 8A). On day 168 and thereafter, LA could be detected but the concentrations could not be determined because they were below the lower

AC

limit of quantification.

The serum concentration profile of LA differed from that observed for the aqueous humor (Fig. 8B). In the serum sample no burst release was observed for LA. Instead, the drug was continuously released over the first 40 weeks after injection reaching a concentration of 13 ng/mL at day 40. Overall, LA concentrations in the serum appeared to be slightly lower than in aqueous humor. In agreement with the LA concentration values observed in aqueous humor, also in the serum the drug content started to decline after 40 days, steadying around 5 ng/mL. After 152 days, LA concentrations in serum were too low for quantification.

ACCEPTED MANUSCRIPT

3.7

AC

Figure 8

CE P

TE

D

MA

NU

SC R

IP

T

15

In vivo degradation, biomicroscopic follow-up

Following implantation of HA/ELA-NCO 1/1 (v/v), the rabbits showed no clinical signs of discomfort. Shortly after injection, we observed a local conjunctival hyperemia. Figure 9B demonstrates the localization of the polymer immediately after application. In this case the deposit presents no vascularity corresponding to MBGS grade 1 in the center and the periphery. After 1 month, a swelling with conjunctival hyperemia was observed biomicroscopically (Fig. 9C), corresponding to MBGS grade 5 in the center and grade 4 in the periphery. Chemosis and vascularity, corresponding to MBGS 7 in the center and MBGS 8-9 in the periphery, were persistent at the injection site for up to 3 months,. (Fig. 9D, E). After 6 months, biomicroscopy revealed a continuous loss of implant mass (Fig.10F) with reduction of vascularity to grade 5 and 6 in the center and the periphery, respectively. After 8 months, only a minimal pigmented subconjunctival lesion was visible (Fig. 9G). The vascularisation regressed completely to grade 1 and no conjunctival scarring or fibrosis was

ACCEPTED MANUSCRIPT 16 seen biomicroscopically at that time point. Over period of 10 months the subconjunctival

3.8

AC

Figure 9

CE P

TE

D

MA

NU

SC R

IP

T

deposit degraded completely (Fig. 9H).

In vivo biocompatibility, histopathological follow-up

One month after injection, the polymer was localized in the subconjunctival connective tissue (Fig. 10A), surrounded by a large wall of lymphocytes, eosinophilic granulocytes, and polynuclear giant cells, demonstrating the tissue response to the components of the LDD system (Fig.10B, C), corresponding to grade IV. A fibrous capsule was not observed at any time point. The inflammatory cells around the LDD system were also present at the 2-month follow-up (Fig. 10D, E, grade IV).

ACCEPTED MANUSCRIPT

TE

D

MA

NU

SC R

IP

T

17

CE P

Figure 10

Fragmentation and beginning resorption of the polymer was observed as early as 1 month (Fig. 10A) and continued steadily up to 9 months after surgery (Figs 10D and 11A, C, E). The neighboring sclera as well as the limbus were not involved in the inflammatory processes,

AC

except in 1 case after 3 months (Fig. 11A, grade IV). Small septa containing lymphocytes and foreign body giant cells, which were observed in all specimens, contributed to the fragmentation and degradation of the polymer (Figs. 11B, D, F). However, no significant fibrous capsule was formed around the foreign body during the observation period (Fig. 11A through F). After 10 months of observation no remaining polymer was visible in the histological section and only very few inflammatory cells were present at the supposed site of former polymer location (Fig. 11G, H, grade I).

ACCEPTED MANUSCRIPT

AC

Figure 11

CE P

TE

D

MA

NU

SC R

IP

T

18

4. Discussion

Glaucoma is the leading cause of irreversible blindness worldwide. It is a heterogeneous set of diseases with regard to pathogenesis, yet increased intraocular pressure (IOP) is associated with most cases of irreversible vision loss. Of the different glaucoma types primary open-angle glaucoma (POAG) is the most common form, resulting from impaired outflow of the aqueous humor through the trabecular meshwork [28]. Current first-line therapy, using topical eye drops designed to lower IOP, is greatly impaired by ocular discomfort and poor patient adherence [29]. These drawbacks have driven research to develop alternative strategies for sustained drug delivery, with subconjunctival injectable LDD systems among them [30]. They offer a promising way to overcome limitations of topical administration, improving patient compliance and therapy outcomes. They may continuously deliver therapeutically effective amounts of drugs to the target site over a prolonged period of

ACCEPTED MANUSCRIPT 19 time, ideally 3 to 4 months. A survey of glaucoma patients has shown that 74% of patients would replace topical application of eye drops with subconjunctival injection [31]. Injections or implantations in the subconjunctival space are minimally invasive and well tolerated by patients. Furthermore, subconjunctivally applied drugs have to penetrate through the sclera,

T

a structure with much better permeability than the cornea. Compared to the cornea where

IP

the junctional complexes between epithelial cells hinder the transport of hydrophilic

SC R

compounds, the scleral tissue, composed of collagen fibers and mucopolysaccharides, has shown a 10 times greater permeability [32]. The higher scleral permeability compared to the cornea could be confirmed for both hydrophilic and hydrophobic compounds [33]. The safety and efficacy of subconjunctival implants has been demonstrated in numerous

NU

animal trials, many of them performed in rabbits. In general, drug release duration shorter than the desired 3 to 4 months was reported. In a short-term study, latanoprost acid-loaded

without

any

adverse

effects

MA

nanoparticles provided in vivo drug delivery and IOP-reduction for 8 consecutive days on

surrounding

tissue

[34].

Chitosan/gelatin/beta-

glycerophosphate-gels loaded with LA demonstrated a continuous drug release over 1 month

D

[35]. Egg-phospatidylcholine-based lyposomes loaded with LA led to an IOP reduction in the subconjunctivally injected eyes for up to 90 days, even though the study did not demonstrate

TE

the in vivo drug release [17]. A normal biocompatibility and drug release for up to 60 days has been demonstrated after subconjunctival injection of poly(-caprolactone) (PCL)-based

CE P

disks loaded with dorzolamide hydrochloride [16]. The longest in vivo release in rabbit eyes (90 days) was demonstrated after subconjunctival implantation of timolol maleate-loaded microspheres composed of PLGA and PLA [36]. However, there are no studies reporting an

AC

LA release in vivo for a longer period than 1 month. Within this context we aimed to develop a polymer-based LDD system enabling a sustained drug release over a prolonged period of time. This injectable system consists of 2 components, HA and ELA-NCO. HA and its derivatives are well-established hydrophilic drug carriers [37,38]. For the presented application HA was chosen, since HA is commonly used in addition to cellulose derivatives as a viscoelastic biomaterial for routine cataract surgery, serving to stabilize the anterior chamber. Furthermore, the potential of HA loaded with antiproliferative model drugs was studied regarding the effectiveness of local drug delivery for lens epithelial cell ablation from the basal membrane and for prevention of secondary cataract formation [39]. However, in that study pure, non-crosslinked hydrogel was used, as only 2-5 minutes of drug release were needed. When used as an LDD system for long-term release, the swelling properties of the hydrogel have to be greatly reduced, which can be achieved by crosslinking HA with ELA-NCO. ELA-NCO is a hydrophobic isocyanate functionalized 1,2-ethylene glycol bis(dilactic), which has been previously investigated as part of a drug eluting tissue adhesive formula [19,22,40]. Micro-CT measurements showed

ACCEPTED MANUSCRIPT 20 that HA/ELA-NCO has a highly porous structure, which is advantageous for drug release and leads to softness. Pores are also beneficial for tissue sprouting and degradation. This was already shown for ELA-NCO in combination with chitosan chloride used as tissue adhesive [22].

T

Previously, the biocompatibility of LDD components has been separately demonstrated for

IP

both, HA and ELA-NCO [19,41]. In the present study, the HA/ELA-NCO mixtures yielded

SC R

likewise good results. The best in vitro biocompatibility could be demonstrated for the 1/1 HA/ELA-NCO (v/v) mixture, which was therefore used in animal experiments. Upon injection, there was no sign of fibrotic capsule formation or pronounced vascularization at any observation time point. The high cellularity represented by occurrence of numerous

NU

lymphocytes and foreign body giant cells at early time points declined gradually. After 10 months, biodegradation seemed to be completed and no resting inflammatory response

MA

could be detected. The extent to which the phagocytosis by macrophages and local release of inflammatory cellular enzymes contributed to polymer degradation remains unclear. Among glaucoma drugs, LA was chosen for drug release experiments, because, as already

D

mentioned, the current literature lacks any data concerning long-term subconjunctival LA delivery. Our preliminary in vivo drug release measurements revealed a sustained LA

TE

delivery over 5 months, with LA concentrations in the aqueous humor ranging from 5 to 20 ng/mL. To our knowledge, this is the longest LA release duration which could be

CE P

established after subconjunctival injection. Notably, an initial burst release (179 ng/mL) was observed within the first 24 hours after application. The burst phenomenon is characteristic for many polymeric drug delivery systems, and as in our case, is often undesired because it

AC

potentially hinders long term release of adequate drug levels [42]. It should be pointed out, however, that in our study we only measured the concentration of the LA ester prodrug, but not the free acid, which is the active compound. It is well recognized that during permeation of LA ester through the cornea the prodrug is hydrolyzed to the free acid by corneal esterases [43]. The active acid form in concentrations between 2 to 30 ng/mL could be detected by Sjöquist and Stjernschantz 24 h after topical LA application [43]. Scleral tissue is likewise able to hydrolyze LA to the active acid form, albeit with a lesser efficiency [44]. Systemic presence of latanoprost free acid after subconjunctival administration of LA-ester has been described previously [45], as well as the passage of latanoprost acid from subconjunctival space into the aqueous humor [34]. We detected LA in aqueous humor and blood serum. The burst release of LA seen in aqueous humor is indicative of a transcleral diffusion from the polymer into the anterior chamber. Given the high concentrations of LA we found in serum, we assume that a systemic absorption from the injection site occurs as well. Systemic levels of LA are also influenced by clearance of LA from the anterior chamber via aqueous humor outflow through

ACCEPTED MANUSCRIPT 21 the trabecular meshwork into the Schlemm’s canal and through the uveoscleral pathway [46]. The precise contribution of ocular diffusion and systemic distribution to the LA concentrations in aqueous humor remains to be elucidated. An important issue when using the biodegradable polymers in LDDs is the degradation time,

T

which should correspond to the drug release duration. We were able to detect and quantify

IP

LA for 152 days in aqueous humor and serum. Thereafter, only detection but no

SC R

quantification of LA was possible in the samples, than the LA concentration was lower than limit of quantification (5 ng/ml) However, degradation of the polymer was not completed for another 4 months, as seen in the histology studies. Whether the amounts of latanoprost released during those 4 months are within the therapeutic range remains to be determined.

NU

The correlation of drug release and degradation in vivo must be addressed by careful tailoring of HA and ELA-NCO to shorten the degradation time and decrease the initial burst

MA

release to enable extended drug delivery. 5. Conclusion

D

In conclusion, we have demonstrated the biocompatibility, feasibility, and desirable drug release profiles of an LDD system based on HA cross-linked by ELA-NCO. Our results

TE

suggest that the LA-loaded, biodegradable, polymer-based LDD system may have potential applications in glaucoma therapy, especially in patients incompliant for various reasons.

CE P

Future experiments with a larger animal cohort are planned to evaluate the concentration of the active drug (latanoprost acid) in aqueous humor after subconjunctival injection. Additionally, investigations will be conducted with the main focus on matching drug release

AC

and degradation time.

ACKNOWLEDGMENTS This study was financially supported by BMBF (Bundesministerium für Bildung und Forschung), within REMEDIS “Höhere Lebensqualität durch neuartige Mikroimplantate” [FKZ: 03IS2081). The authors thank Ms. Claudia Lurtz for her assistance in µ-CT measurements and Prof. Gerhard Hennighausen for his helpful notes and suggestions. The excellent histological work of Ms. Heike Brückmann is greatly appreciated.

ACCEPTED MANUSCRIPT 22 REFERENCES

N. Pfeiffer, J. Lamparter, A. Gericke, F.H. Grus, E.M. Hoffmann, J. Wahl, Neuroprotection of medical IOP-lowering therapy, Cell and tissue research 353 (2013) 245–251.

[2]

A. Heijl, C.M. Leske, B. Bengtsson, L. Hyman, B. Bengtsson, M. Hussein, Reduction of Intraocular Pressure and Glaucoma Progression, Archives of Ophthalmology 120 (2002) 1268–1279.

[3]

R.E. Marquis, J.T. Whitson, Management of Glaucoma: Focus on Pharmacological Therapy, Drugs & Aging 22 (2005) 1–21.

[4]

S. Koevary, Pharmacokinetics of Topical Ocular Drug Delivery: Potential Uses for the Treatment of Diseases of the Posterior Segment and Beyond, Current Drug Metabolism 4 (2003) 213–222.

[5]

A.P. Rotchford, K.M. Murphy, Compliance with timolol treatment in glaucoma, Eye 12 (Pt 2) (1998) 234–236.

[6]

G. Reardon, S. Kotak, G.F. Schwartz, Objective assessment of compliance and persistence among patients treated for glaucoma and ocular hypertension: a systematic review, Patient preference and adherence 5 (2011) 441–463.

[7]

B. Sleath, S. Blalock, D. Covert, J.L. Stone, A.C. Skinner, K. Muir, A.L. Robin, The relationship between glaucoma medication adherence, eye drop technique, and visual field defect severity, Ophthalmology 118 (2011) 2398–2402.

[8]

C. Baudouin, A. Labbé, H. Liang, A. Pauly, F. Brignole-Baudouin, Preservatives in eyedrops: the good, the bad and the ugly, Progress in retinal and eye research 29 (2010) 312–334.

[9]

W.C. Stewart, J.A. Stewart, L.A. Nelson, Ocular surface disease in patients with ocular hypertension and glaucoma, Current eye research 36 (2011) 391–398.

CE P

TE

D

MA

NU

SC R

IP

T

[1]

[10] D.E. Everitt, J. Avorn, Systemic Effects of Medications Used To Treat Glaucoma, Ann Intern Med 112 (1990) 120–125.

AC

[11] E. Lavik, M.H. Kuehn, Y.H. Kwon, Novel drug delivery systems for glaucoma, Eye 25 (2011) 578–586. [12] A. Ludwig, The use of mucoadhesive polymers in ocular drug delivery, Advanced drug delivery reviews 57 (2005) 1595–1639. [13] S.S. Anumolu, Y. Singh, D. Gao, S. Stein, P.J. Sinko, Design and evaluation of novel fast forming pilocarpine-loaded ocular hydrogels for sustained pharmacological response, Journal of Controlled Release 137 (2009) 152–159. [14] J.B. Ciolino, C.F. Stefanescu, A.E. Ross, B. Salvador-Culla, P. Cortez, E.M. Ford, K.A. Wymbs, S.L. Sprague, D.R. Mascoop, S.S. Rudina, S.A. Trauger, F. Cade, D.S. Kohane, In vivo performance of a drug-eluting contact lens to treat glaucoma for a month, Biomaterials 35 (2014) 432–439. [15] D. Manickavasagam, M.O. Oyewumi, Critical assessment of implantable drug delivery devices in glaucoma management, Journal of drug delivery 2013 (2013) 895013. [16] M.V. Natu, M.N. Gaspar, C.A. Fontes Ribeiro, A.M. Cabrita, H.C. de Sousa, M.H. Gil, In vitro and in vivo evaluation of an intraocular implant for glaucoma treatment, International journal of pharmaceutics 415 (2011) 73–82.

ACCEPTED MANUSCRIPT 23 [17] J.V. Natarajan, M. Ang, A. Darwitan, S. Chattopadhyay, T.T. Wong, S.S. Venkatraman, Nanomedicine for glaucoma: liposomes provide sustained release of latanoprost in the eye, International journal of nanomedicine 7 (2012) 123–131.

T

[18] K.L. Goa, P. Benfield, Hyaluronic acid. A review of its pharmacology and use as a surgical aid in ophthalmology, and its therapeutic potential in joint disease and wound healing, Drugs 47 (1994) 536–566.

SC R

IP

[19] H.W. Rohm, C. Lurtz, J. Wegmann, E.K. Odermatt, D. Behrend, K.-P. Schmitz, K. Sternberg, Development of a biodegradable tissue adhesive based on functionalized 1,2-ethylene glycol bis(dilactic acid). II, Journal of biomedical materials research B: Applied biomaterials 97B (2011) 66–73.

NU

[20] K. Sternberg, K. Voss, C. Kastner, A. Bernsdorf, C. Lurtz, O. Stachs, R.F. Guthoff, K.P. Schmitz, Development of an Injectable Polymer-Based Local Drug Delivery System for Subconjunctival Treatment of Glaucoma, Biomed Tech 58 (Suppl. 1) (2013) 147– 148.

MA

[21] C. Heiss, N. Hahn, S. Wenisch, V. Alt, P. Pokinskyj, U. Horas, O. Kilian, R. Schnettler, The tissue response to an alkylene bis(dilactoyl)-methacrylate bone adhesive, Biomaterials 26 (2005) 1389–1396. [22] K. Sternberg, H.W. Rohm, C. Lurtz, J. Wegmann, E.K. Odermatt, D. Behrend, D. Michalik, K.-P. Schmitz, Development of a biodegradable tissue adhesive based on functionalized 1,2-ethylene glycol bis(dilactic acid). I, Journal of biomedical materials research. Part B, Applied biomaterials 94 (2010) 318–326.

TE

D

[23] T. Freier, H.S. Koh, K. Kazazian, M.S. Shoichet, Controlling cell adhesion and degradation of chitosan films by N-acetylation, Biomaterials 26 (2005) 5872–5878.

CE P

[24] T. Stahnke, M. Löbler, C. Kastner, O. Stachs, A. Wree, K. Sternberg, K.-P. Schmitz, R. Guthoff, Different fibroblast subpopulations of the eye: a therapeutic target to prevent postoperative fibrosis in glaucoma therapy, Experimental eye research 100 (2012) 88– 97. [25] A.P. Wells, J.G. Crowston, J. Marks, J.F. Kirwan, G. Smith, Clarke, J C K, R. Shah, J. Vieira, C. Bunce, I. Murdoch, P.T. Khaw, A pilot study of a system for grading of drainage blebs after glaucoma surgery, Journal of glaucoma 13 (2004) 454–460.

AC

[26] L. Prantl, S. Schreml, S. Fichtner-Feigl, N. Pöppl, M. Eisenmann-Klein, H. Schwarze, B. Füchtmeier, Clinical and morphological conditions in capsular contracture formed around silicone breast implants, Plastic and reconstructive surgery 120 (2007) 275– 284. [27] Thompson, Lester D R, Pheochromocytoma of the Adrenal gland Scaled Score (PASS) to separate benign from malignant neoplasms: a clinicopathologic and immunophenotypic study of 100 cases, The American journal of surgical pathology 26 (2002) 551–566. [28] M. Johnson, 'What controls aqueous humour outflow resistance?', Experimental eye research 82 (2006) 545–557. [29] I. Januleviciene, L. Siaudvytyte, R. Barsauskaite, Ophthalmic drug delivery in glaucoma-a review, Pharmaceutics 4 (2012) 243–251. [30] N. Gooch, S.A. Molokhia, R. Condie, R.M. Burr, B. Archer, B.K. Ambati, B. Wirostko, Ocular drug delivery for glaucoma management, Pharmaceutics 4 (2012) 197–211. [31] R.S. Chong, Su, Daniel H W, A. Tsai, Y. Jiang, H.M. Htoon, E.L. Lamoureux, T. Aung, T.T. Wong, Patient acceptance and attitude toward an alternative method of subconjunctival injection for the medical treatment of glaucoma, Journal of glaucoma 22 (2013) 190–194.

ACCEPTED MANUSCRIPT 24 [32] K.M. Hämäläinen, K. Kananen, S. Auriola, K. Kontturi, A. Urtti, Characterization of paracellular and aqueous penetration routes in cornea, conjunctiva, and sclera, Investigative ophthalmology & visual science 38 (1997) 627–634.

T

[33] M.R. Prausnitz, J.S. Noonan, Permeability of cornea, sclera, and conjunctiva: a literature analysis for drug delivery to the eye, Journal of pharmaceutical sciences 87 (1998) 1479–1488.

SC R

IP

[34] A. Giarmoukakis, G. Labiris, H. Sideroudi, Z. Tsimali, N. Koutsospyrou, K. Avgoustakis, V. Kozobolis, Biodegradable nanoparticles for controlled subconjunctival delivery of latanoprost acid: in vitro and in vivo evaluation. Preliminary results, Experimental eye research 112 (2013) 29–36. [35] Y.-H. Cheng, K.-H. Hung, T.-H. Tsai, C.-J. Lee, R.-Y. Ku, A.W.-H. Chiu, S.-H. Chiou, C.J.-L. Liu, Sustained delivery of latanoprost by thermosensitive chitosan-gelatin-based hydrogel for controlling ocular hypertension, Acta biomaterialia 10 (2014) 4360–4366.

NU

[36] J.P. Bertram, S.S. Saluja, J. McKain, E.B. Lavik, Sustained delivery of timolol maleate from poly(lactic-co-glycolic acid)/poly(lactic acid) microspheres for over 3 months, Journal of microencapsulation 26 (2009) 18–26.

MA

[37] T. Hirakura, K. Yasugi, T. Nemoto, M. Sato, T. Shimoboji, Y. Aso, N. Morimoto, K. Akiyoshi, Hybrid hyaluronan hydrogel encapsulating nanogel as a protein nanocarrier: new system for sustained delivery of protein with a chaperone-like function, Journal of Controlled Release 142 (2010) 483–489.

D

[38] J.R. Joshi, R.P. Patel, Role of biodegradable polymers in drug delivery, International J Current Pharma Research 4 (2012) 74–81.

TE

[39] K. Sternberg, T. Terwee, O. Stachs, R. Guthoff, M. Löbler, K.-P. Schmitz, Drug-induced secondary cataract prevention: experimental ex vivo and in vivo results with disulfiram, methotrexate and actinomycin D, Ophthalmic research 44 (2010) 225–236.

CE P

[40] C. Lurtz, K. Voss, V. Hahn, F. Schauer, J. Wegmann, E.K. Odermatt, K.-P. Schmitz, K. Sternberg, In vitro degradation and drug release of a biodegradable tissue adhesive based on functionalized 1,2-ethylene glycol bis(dilactic acid) and chitosan, Journal of materials science. Materials in medicine 24 (2013) 667–678.

AC

[41] K. Miyamoto, M. Sasaki, Y. Minamisawa, Y. Kurahashi, H. Kano, S.-i. Ishikawa, Evaluation of in vivo biocompatibility and biodegradation of photocrosslinked hyaluronate hydrogels (HADgels), Journal of biomedical materials research. Part A 70 (2004) 550–559. [42] X. Huang, C.S. Brazel, On the importance and mechanisms of burst release in matrixcontrolled drug delivery systems, Journal of Controlled Release 73 (2001) 121–136. [43] B. Sjöquist, J. Stjernschantz, Ocular and Systemic Pharmacokinetics Of Latanoprost in Humans, Survey of Ophthalmology 47 (2002) S6. [44] F.L. Guerra, J. Rager, S. Bhoopathy, I. Hidalgo, K. Castermans, O. Defert, S. Boland, In Vitro Hydrolysis of Latanoprost by Human Ocular Tissues, Invest. Ophthalmol. Vis. Sci. 53 (2012) 5319. [45] B.A. Jessen, Shiue, Michael H I, H. Kaur, P. Miller, R. Leedle, H. Guo, M. Evans, Safety assessment of subconjunctivally implanted devices containing latanoprost in Dutch-belted rabbits, Journal of ocular pharmacology and therapeutics 29 (2013) 574– 585. [46] M. Goel, R.G. Picciani, R.K. Lee, S.K. Bhattacharya, Aqueous humor dynamics: a review, The open ophthalmology journal 4 (2010) 52–59.

ACCEPTED MANUSCRIPT 25 LEGENDS Figure 1: Polymerization scheme of ELA-NCO with HA. The possible sites for reaction are

T

shown exemplarily.

IP

Figure 2: FTIR-ATR spectra of HA/ELA-NCO 1/1 (v/v) at room temperature after reaction

SC R

times of 2 h, 4 h, 7 h, 24 h and 48 h.

Figure 3:

Cumulative drug release of latanoprost (LA) from HA(LA)/ELA-NCO LDD systems in 0.9%

NU

NaCl solution at 37 °C. Data are presented as mean values (n=6) ± SD.

MA

Figure 4: In vitro degradation-induced mass loss of HA/ELA-NCO mixtures (1/1, 1/4 and 1/10 (v/v)) at 37 °C in comparison to the reference material poly(L-lactide) (PLLA). Data are

D

presented as mean values (n=6) ± SD.

Figure 5: Macrophotographic documentation of the in vitro degradation-induced changes in

TE

sample diameter (A) and thickness (B) of HA/ELA-NCO (1/1 (v/v)) after 26 weeks. C displays processed images (grayscale) for 3D-integrated µ-CT analyses of the

CE P

HA/ELA-NCO sample porosities at baseline (t = 0) and following 26 weeks in vitro degradation (t = 26 weeks).

AC

Figure 6: Relative viability of human Tenon’s fibroblasts incubated with different dilutions of ELA-NCO eluate (A) and HA/ELA-NCO eluates (B, C and D for 1/10, 1/4 and 1/1, respectively). The values are presented as mean of 4 independent experiments ± SEM. Figure 7: Relative viability of rabbit Tenon’s fibroblasts incubated with different dilutions of ELA-NCO eluate (A) and HA/ELA-NCO eluates (B, C and D for 1/10, 1/4 and 1/1, respectively). The values are presented as mean of 4 independent experiments ± SEM.

Figure 8: Time course of LA release in vivo. Concentration of LA in aqueous humor (A) and serum (B) obtained via HPLC-MS after subconjunctival application of approximately 50 mg HA(LA)/ELA-NCO (1/1 v). * indicates the time point 15 minutes after injection. Figure 9: Macroscopic documentation of the injection site before (A) and after subconjunctival injection of approximately 50 mg HA/ELA-NCO (1/1 (v/v)) (B through H).

ACCEPTED MANUSCRIPT 26 HA/ELA-NCO appearance direct after injection (B) and at different time points following injection: C–1 month, D–2 months, E–3 months, F–6 months, G–8 months, H-10 months.

Figure 10: Histological evaluation of polymer injected into the subconjunctival connective

T

tissue and its appearance after a period of 2 months (hematoxilin/eosin staining).

IP

A: One month after injection. The remnants of the polymer (arrows) are surrounded by a

SC R

heavy leuco-lymphatic infiltration. The sclera (Sc) is free. The region marked by a rectangle is shown in B. Scale bar: 250 µm.

B: Magnification of the area displayed in A. Fragment of the polymer surrounded by non-

polymer (arrowheads). Scale bar: 50 µm.

NU

organized inflammatory cells. A giant cell (arrow) has generated a resorption vacuole of the

C: A polynuclear giant cell (arrow) enwrapping a polymer fragment. Many eosinophilic

MA

granulocytes were located nearby. Scale bar: 10 µm.

D: Two months after injection. The polymer was surrounded by a dense wall of inflammatory cells. Scale bar: 500 µm.

D

E: A cavity with partially resorbed material was still surrounded by some giant cells (arrows).

TE

Scale bar: 50 µm.

Figure 11: Histological evaluation of polymer resorption (continued): 3-10 months. Regions

CE P

in the left column marked by a rectangle are magnified in the right column. A: Three months after injection. Severe inflammatory reaction around the polymer injection site with partial infiltration of the sclera (arrowheads). Small septa organize compartments at

AC

the margins of a large polymer body (arrows). Scale bar: 500 µm. B: A cross-sectioned septum containing epitheloid cells and with polynuclear giant cells (arrows). Increased formation of capillaries. Scale bar: 50 µm. C: Eight months after injection. The polymer is still present, but the lymphocytic wall appears reduced in size. Scale bar: 500 µm. D: Higher magnification of the situation shown in C. Small septa containing lymphocytes enwrap the polymer. Giant cells are marked by arrows. Scale bar: 50 µm. E: Nine months after injection. In this animal, the injection site and the polymer does not appear to be reduced compared to earlier stages. Scale bar: 500 µm. F: Magnification for E. Advanced fragmentation of the polymer, however, with many giant cells and eosinophilic granulocytes. Scale bar: 25 µm. G: Ten months after injection. In this animal, neither polymer nor noteworthy numbers of inflammatory cells were found, possibly somewhat edematous connective tissue, compared to the nasal control side of the conjunctiva (not shown). Scale bar: 100 µm.

ACCEPTED MANUSCRIPT 27 H: Magnification for G. Empty spaces with some septum-like tongues of adjacent connective tissue may point to the site of former polymer location. Scale bar: 25 µm.

AC

CE P

TE

D

MA

NU

SC R

IP

T

Table 1: Modified Wilfingseder grading system

ACCEPTED MANUSCRIPT 28 Chemicals PubChem CID: 174

Lactide (3,6-dimethyl-1,4-dioxane-2,5-dione):

PubChem CID: 7272

Phosphoric acid:

PubChem CID: 1004

Hexamethylene diisocyanate:

PubChem CID: 13192

Dimethylsulfoxide:

PubChem CID: 679

Poly(l-lactide) (D-lactic acid?):

PubChem CID: 61503

Latanoprost:

PubChem CID: 5311221

Egg white lysozyme (19-36) [Gallus gallus]:

PubChem CID: 71581463

Hyaluronic acid sodium salt:

PubChem CID: 53398704

Sodium chloride:

PubChem CID: 5234

NU

SC R

IP

T

Ethylene glycol:

DMEM: Dulbecco's Modified Eagle's Medium:

PubChem SID: 56312938

with

PubChem CID: 5793

MA

D(+)-Glucose: L-Glutamine:

Sodium chloride; isotonic: Sodium chloride:

D

with

CE P

Penicillin/Streptomycin Pen-Strep:

AC

with

TE

Penicillin/Streptomycin Pen-Strep:

PubChem CID: 5961 PubChem CID: 103770557 PubChem CID: 5234 PubChem SID: 162248774 PubChem CID: 71311919

ACCEPTED MANUSCRIPT 29

AC

CE P

TE

D

MA

NU

SC R

IP

T

Graphical Abstract