Treatment Efficacy of a Dual Release of Aflibercept and Dexamethasone From a Single Hydrogel Drug Delivery System in a Rodent Model

Kayla M. Rudeen; Chryssa M. Maloney; Katherine L. Lydon; Leandro B. C. Teixeria; William F. Mieler; Jennifer J. Kang-Mieler

doi:10.1167/tvst.14.6.31

Abstract

Purpose: Age-related macular degeneration (AMD) is the leading cause of vision loss for the elderly population. Wet AMD, which accounts for approximately 15% of AMD cases, is characterized by abnormal blood vessel growth from the choroid into the subretinal space. Although intravitreal anti-vascular endothelial growth factor (VEGF) agents have become the standard of care for wet AMD, there is a growing subset of patients who do not fully respond to monotherapy anti-VEGF treatment. In previously published reports, corticosteroids have shown improvements in treatment efficacy when administered with anti-VEGF in a subset of non-responders to anti-VEGF monotherapy.

Methods: A combination dexamethasone and aflibercept drug delivery system (Combo-DDS) was evaluated in a laser-induced rodent model of choroidal neovascularization (CNV). Longitudinal monitoring was done through week 22 using fluorescein angiography (FA) and spectral domain optical coherence tomography (SD-OCT). Multi-Otsu thresholding was used to quantify the lesion area based on late-phase FA images. In addition, preliminary safety and biocompatibility of the Combo-DDS were evaluated by intraocular pressure (IOP) measurements, electroretinogram (ERG), and histology (n = 6 eyes/group).

Results: In the laser-induced CNV model, CNV lesions (n = 28–36 lesions/group) were monitored longitudinally. Combo-DDS showed a regression in lesion size starting at week 2 that continued through the end of study. IOP, ERG, and histology showed preliminary safety and biocompatibility of the Combo-DDS.

Conclusions: This study demonstrated that Combo-DDS maintained treatment efficacy in a laser-induced CNV rodent model for 6 months.

Translational Relevance: The Combo-DDS shows the potential to eliminate the need for separate dosing regiments of anti-VEGF and corticosteroids for non-responders to anti-VEGF monotherapy.

Introduction

Age-related macular degeneration (AMD) is the leading cause of vision loss for the elderly population.¹^,² It is estimated that 196 million people have AMD worldwide, and that number is expected to increase to 288 million by 2040.³ There are two types of AMD: dry (geographic) and wet (neovascular) AMD. Approximately 85% of those patients are diagnosed with dry AMD, which is characterized by focal deposits of debris, known as drusen, under the macula, or by geographic atrophy of the retinal pigment epithelium cells, photoreceptors, and choriocapillaris. Wet AMD includes macular neovascularization (MNV), which is abnormal blood vessel growth beneath the macular. Choroidal neovascularization (CNV) secondary to AMD, which presents through both inflammatory and angiogenesis pathways, may result in rapid vision loss from leakage, exudation, and hemorrhage in the macula.³^–⁵ Therefore, wet AMD tends to be much more aggressive and severe than dry AMD due to the potential rapid deterioration of the macula from MNV.

The pathogenesis of CNV is not well known in patients with AMD, but key studies have shown that vascular endothelial growth factor (VEGF) agents are a major factor in mediating intraocular neovascularization and increased vascular permeability.⁶^–¹⁰ Thus, anti-VEGF agents have become the gold standard of care for wet AMD. Aflibercept (Eylea; Regeneron, Tarrytown, NY, USA) is a highly administered anti-VEGF due to its relatively high binding affinity, half-life, and longer time between injections. Aflibercept is intravitreally administered every 4 weeks for the first 3 months, then every 8 weeks afterward, although treatment regimens do vary.¹¹

Despite the improvement of anti-VEGF therapies for wet AMD, 20% to 40% of patients with wet AMD do not fully respond to monotherapy anti-VEGF treatment.¹²^,¹³ These patients may continue to have persistent fluid despite 2 years of monthly anti-VEGF treatment.¹⁴ This subset of patients is referred to as anti-VEGF non-responders, but there is no universally agreed upon definition for what constitutes a non-responder. It is not entirely known why these patients do not respond, but there have been some indications that underlying genetic differences may cause irregularities in response.¹⁴

Corticosteroids, including triamcinolone and dexamethasone, have shown improvements in treatment efficacy when administered with anti-VEGF in a subset of non-responders to anti-VEGF monotherapy.¹⁴^–¹⁷ The addition of corticosteroids targets the inflammatory pathway of CNV that is not addressed by anti-VEGF agents.¹⁸^–²² Intravitreal (IVT) corticosteroids have some potential side effects, including increase in intraocular pressure (IOP), posterior cataract formation, and endophthalmitis; it is also unclear if corticosteroids provide long-term visual stabilization.²⁰ Despite these side effects, non-responders may benefit from combination therapy of corticosteroids and anti-VEGF therapy.¹⁶^,¹⁷ This combination treatment would require separate dosing regiments for each drug: monthly or bi-monthly injections of anti-VEGFs with quarterly or semi-annual injections of a dexamethasone implant such as Ozurdex (Allergan, an AbbVie company, Irvine, CA, USA).

Our laboratory has developed a drug delivery system (DDS) that is able to control and extend the release of aflibercept for 6 months.²³^–²⁷ It is a composite system of aflibercept-loaded poly(D,L-lactide-co-glycolide) acid (PLGA) microspheres embedded into a biodegradable, thermoresponsive poly(ethylene glycol)-co-(L-lactic acid) diacrylate (PEG-PLLA-DA)-N-isopropylacrylamide (NIPAAm) hydrogel. The volume phase transition temperature of the hydrogel is below physiological temperatures (32.5°C), which allows for the DDS to be injected through a 28-gauge needle at room temperature then further solidify upon reaching physiological temperature in the vitreous cavity of the eye.²³^,²⁴ Additionally, the bioactivity of aflibercept released from the DDS was confirmed through enzyme-linked immunosorbent assay (ELISA), dot blot assay, and MTS proliferation assay.²³^,²⁴^,²⁷ It has shown similar treatment efficacy to aflibercept bolus IVT injections in a laser-induced CNV model in rodents for 6 months,²⁵ as well as safety and biocompatibility in non-human primates (NHP).²⁶ To minimize the number of corticosteroids and anti-VEGF combination injections to treat monotherapy non-responders, we modified our DDS to simultaneously release aflibercept (AFL) and dexamethasone (DEX) for 6 months by embedding both biodegradable microparticles and nanoparticles into a single hydrogel.²⁸ Previously in vitro characterization of this system showed that extended and controlled the release could be achieved of both aflibercept for approximately 6 months.²⁸ This combination-DDS (Combo-DDS) may eliminate the need for separate dosing regiments and reduce the total number of injections for non-responsive patients to monotherapy.

The purpose of this study was to evaluate if Combo-DDS (DEX and AFL) has similar treatment efficacy compared to Bolus AFL and AFL-DDS in a rodent CNV model. Longitudinal monitoring was performed using fluorescein angiography (FA) and spectral domain optical coherence tomography (SD-OCT). Multi-Otsu thresholding was used to quantify lesion area based on late-phase FA images.²⁵^,²⁸^,²⁹ In addition, preliminary safety and biocompatibility of the Combo-DDS were evaluated by intraocular pressure (IOP) measurements, electroretinogram (ERG), and histology.

Methods

Animal Preparation

All animal procedures were in accordance with protocols approved by the Institutional Animal Care and Use Committee at the Illinois Institute of Technology and with the principles embodied in the statement on the use of animals in ophthalmic and vision research adopted by the Association for Research in Vision and Ophthalmology (ARVO).

Long-Evans male rats (200–250 g) were purchased from Envigo Laboratories (Indianapolis, IN, USA). Animals were anesthetized using 80 mg/kg of ketamine hydrochloride (Henry Schein Animal Health, Dublin, OH, USA) and 10 mg/kg xylazine (Henry Schein Animal Health) via intraperitoneal (IP) injection. Proparacaine drops (Bausch and Lomb, Rochester, NY, USA) were used to anesthetize the corneas, followed by both phenylephrine (Bausch and Lomb) and atropine drops (Bausch and Lomb) to dilate the pupils. Blood oxygen saturation and heart rate were monitored with a pulse oximeter (8500 AV; Nonin Medical Inc., Plymouth, MN, USA). Animals were placed on a heated stage and monitored to maintain a core body temperature of 37°C during all procedures.

Preparation of Sterile DDS

Dexamethasone (Sigma-Aldrich, St. Louis, MO, USA) was encapsulated into PLGA (75:25, MW 4–15 kDa [kilodalton]) nanoparticles by a modified single-emulsion, solvent evaporation technique.³⁰^–³² Briefly, dexamethasone and PLGA were dissolved in dichloromethane (DCM) to create the oil phase, and polyvinyl alcohol (PVA) was used as the water phase. The oil-in-water emulsion (o/w) was created by sonication at 100 watts for 3.5 minutes. After the solvent was allowed to evaporate, dexamethasone loaded-nanoparticles (DEX-nps) were collected by centrifugation, washed with deionized (DI) water at least 3 times, lyophilized, and stored at 4°C. Blank nanoparticles utilized the same technique with the absence of the drug.

Similarly, aflibercept (Eylea; Regeneron, Tarrytown, NY, USA) was encapsulated into PLGA (75:25) microspheres by a modified double-emulsion, solvent evaporation technique.²³^,²⁴^,²⁷^,³⁰ The first emulsion (w1/o) was created by vortex, then immediately added to PVA (w2) to create a double emulsion (w1/0/w2) by vortex. Excipients were added to the inner aqueous phase (w1) for protein stabilization and to the oil phase (o) to act as a buffer.²³^,²⁷ After solvent evaporation, aflibercept loaded-microparticles (AFL-mps) were collected by centrifugation, washed with DI water at least 3 times, lyophilized, and stored at 4°C. Blank microparticles utilized the same technique with the absence of the drug.

Blank particles, AFL-mps, and DEX-nps were placed under ultraviolet light for 30 minutes. Thermoresponsive, biodegradable PEG-PLLA-DA-NIPAAm hydrogels were synthesized by free radical polymerization, described elsewhere.²³^,²⁴^,³⁰^,³³ NIPAAm (350 mM), N-tert-butylacrylamide (50 mM), ammonium persulfate (13 mM), and PEG-PLLA-DA (2 mM) were dissolved into 1× PBS (pH 7.4) to prepare the hydrogel precursor solution.²⁴ Loading doses of particles were suspended into the solution to create the blank-DDS (20 mg/mL of blank microparticles and nanoparticles), DEX-DDS (20 mg/mL of DEX-nps), Combo-DDS (20 mg/mL of DEX-nps and AFL-mps), and ALF-DDS (20 mg/mL of AFL-mps). The hydrogel precursor was prepared under aseptic processing then filtered through a sterile 13 mm filtered syringe (0.22 µm; Fisherbrand, Thermo Fisher Scientific, Waltham, MA, USA). After polymerization, DDSs were washed in PBS 5 times, then loaded into 0.5 cc U-100 insulin syringes (28 gauge; Becton Dickinson & Co., Franklin Lakes, NJ, USA). Loaded syringes were stored in sterile conditions at 4°C until use.

Laser-Induced Model of CNV

Laser photocoagulation was performed using an Argon-green laser (AKC-8000; NIDEK, Inc., Fremont, CA, USA) attached to a slit lamp with a laser power of 550 mW, duration 100 ms, and spot diameter of 50 µm. A 90-diopter lens was used to view the posterior pole of the eye, then the laser beam focused on the retina to induce the lesions. Five to seven lesions per eye were induced two-to-three-disc diameters from, and centered on, the optic disc. Laser-induced disruption of the Bruch's membrane (BM) was identified by the appearance of a bubble at the site of photocoagulation, and later confirmed with SD-OCT. Only laser spots that resulted in a disruption of the membrane and continued to form leaking hyper-fluorescence lesions were included in analysis.

Experimental Design

Eighteen Long-Evans rats were checked for ocular integrity and then were randomly assigned into one of the following treatment groups giving three rats or six eyes per group: control (No Treatment), blank-DDS (5 µL, no drugs), Bolus AFL IVT injections (5 µL, 10 µg, bimonthly), AFL-DDS (5 µL, 1.5 µg), Combo-DDS (5 µL, 1.5 µg AFL, and 200 µg DEX), and DEX-DDS (5 µL, 200 µg). Treatment intervention was administered 2 weeks after laser induction to allow for full formation of CNV lesions.³⁴^,³⁵

Spectralis HRA + OCT system (Heidelberg Engineering, Heidelberg, Germany) was used to image the retina. Late-phase FA images of CNV lesions were captured at 15 minutes after IP injection of 0.4 mL of 10% fluorescein dye (Sigma-Aldrich). Images were acquired prior to CNV induction and at predetermined time points post induction, as shown in Figure 1, which is a detailed schematic for experimental design and timeline for CNV treatment efficacy studies.

Figure 1.

View Original Download Slide

Study design for determining CNV treatment efficacy in a laser-induced CNV model.

CNV Lesion Quantification

The Multi-Otsu Thresholding (MOT) technique was used to analyze FA images and determine the area of CNV lesions.²⁵^,²⁸^,²⁹ A plug-in for MOT was used in ImageJ software (NIH, USA), which automatically separates the image into two to five levels by maximizing the inter-class variance of pixel intensity.²⁸^,²⁹ Three levels will be used for image processing, which are the background, diffuse leakage, and CNV lesion. This technique excludes blood vessels and allows for only the core lesion to be measured in the number of pixels and then converted to area (µm²). FA images from week 0 (14 days after CNV induction) will be used to normalize all other time points and determine percent change of lesion area. Statistical analysis will be performed with 2-way analysis of variance (ANOVA) and Holm-Sidak multiple comparisons tests (GraphPad Prism 9).

Intraocular Pressure

IOP was determined using a tonometer (TONO-PEN XL; Medtronic, Minneapolis, MN, USA). Measurements were conducted pre-injection, post-injection, and at each time point thereafter. The tonometer was calibrated per the manufacturer's manual.

Histopathology

Animals were euthanized using condensed carbon dioxide. Eyes were enucleated, immersion fixed in Davidson's solution for 48 hours, and then transferred to neutral buffered formalin before submission to the Comparative Ocular Pathology Laboratory of Wisconsin (COPLOW) for histopathology. Eyes were sectioned on the sagittal plane and tissues were processed and paraffin embedded. Serial sections, 5 µm in width, were taken every 50 µm through the target lesions. Hematoxylin and eosin (H&E) staining was used to evaluate fibrovascular growth of the lesions by measuring relative retinal thickness. Relative retinal thickness was defined as thickness from the bottom of the choroid to the top of the CNV lesion relative to the thickness of the adjacent, normal choroid.³⁶

Electroretinography

An ERG was used to determine biocompatibility by evaluating the electrophysiology of the retinal in terms of changes in maximum response and sensitivity of a- and b-waves.³⁷^–⁴¹ Long-Evans rats were dark-adapted overnight, and ERG was conducted in scotopic conditions. ERG responses were recorded using a gold electrode placed on the cornea, reference electrode placed subdermally into the cheek, and grounding electrode placed subdermally into the nape. For reference and ground electrodes, 30 G platinum subdermal needle electrodes (SAFELEAD; Grass Products, Natus Neurology, Warwick, RI, USA) were used. A-wave and b-wave amplitudes were recorded after single 2 ms flashes of light with increasing intensity (0.000311 to 311 sc·cd·s·m-2). Different light intensities were achieved by using a series of log unit neutral density filters. All ERGs were amplified (1000×), filtered with low pass (3 kilohertz [kHz]), and digitized (10 kHz). Animals were administered a 5 µL intravitreal injection of Combo-DDS in each eye at week 0. Prior to injection, baseline ERG (control) was conducted. Then, an ERG was conducted longitudinally for 6 months.

Retinal Thickness Measurements

SD-OCT images were used to quantify retinal thickness away from DDS (control) and near DDS. Retinal thickness measurements were used to determine if any signs of ocular toxicity were present after DDS administration.⁴²

Statistical Analysis

For CNV analysis, each lesion was considered independent and was normalized based on their individual areas at week 0. One-way ANOVA and then Holm-Sidak multiple comparison was performed for histology. Two-way ANOVA then Holm-Sidak multiple comparisons was performed for CNV lesions across treatment groups and time points. The Student's t-test was used to evaluate control against later time points for ERG, IOP, and SD-OCT. All values were reported as mean ± standard error.

Results

CNV Lesion Quantification

A total of 194 CNV lesions were included in the study: 30 lesions in the No Treatment group, 28 lesions in the Blank-DDS group, 36 lesions in the Bolus AFL group, 35 lesions in the AFL-DDS group, 33 lesions in the DEX-DDS group, and 32 lesions in the Combo-DDS group. Representative late-phase FA images of each treatment group for week 0 (prior to treatment intervention) and week 22 are shown in Figure 2.

Figure 2.

View Original Download Slide

Representative late-phase imaging of CNV lesions from all treatment groups at weeks 0 and 22.

While qualitative changes in lesion size can be seen from FA images, MOT was used to objectively quantify the lesion area at each time point using late-phase FA images. Number of lesions and average lesion size per treatment group can be seen in Table 1. There were no significant differences for average lesion size between any treatment groups at week 0 (P > 0.5). Lesion sizes at week 0 were used to determine percent growth (or regression) of lesions for all proceeding timepoints, as shown in Figure 3. For DEX-DDS, AFL-DDS, and Combo-DDS, regressions in lesion size were seen for all time points starting at week 2. Fluctuations in lesion size were observed for the No Treatment and the Blank-DDS groups. Last, the Bolus AFL group had an initial increase in lesion growth at week 2 followed by regression that remained until week 22. At the final time point, increases in lesion size were seen in the No Treatment group (1.9 ± 8.2%) and the Blank-DDS group (4.1 ± 9.9%), whereas decreases were seen in the Bolus AFL group (−11.3 ± 6.9%), the AFL-DDS group (−32.9 ± 4.5%), the DEX-DDS group (−12.5 ± 7.4%), and the Combo-DDS group (−30.0 ± 7.4).

Table 1.

View Table

Characterization of CNV Lesions

Figure 3.

View Original Download Slide

Percent changes in CNV lesion area relative to week 0 for all treatment groups at each time point. Experiment groups included: No Treatment (black; n = 30 lesions), Blank-DDS (blue; n = 28 lesions), Bolus AFL (pink; n = 36 lesions), AFL-DDS (green; n = 35 lesions), DEX-DDS (dark purple; n = 35 lesions), and Combo-DDS (light purple; n = 32 lesions). Statistically significant differences were determined between groups using Holm-Sidak multiple comparisons tests after repeated-measures 2-way ANOVA. Statistical differences of P < 0.05 is represented by *, P < 0.01 is represented by **, P < 0.001 is represented by ***, and P < 0.0001 by ****.

Significant differences in lesions size were seen between the Combo-DDS group with No Treatment (weeks 2, 6, 14, and 22), the Blank-DDS group (weeks 6, 14, 18, and 22), the Bolus AFL group (weeks 14 and 18), and the DEX-DDS group (week 14). Similarly, AFL-DDS had significant differences in lesion size compared to the No Treatment group (weeks 2, 14, and 22), the Blank-DDS group (week 22), and the Bolus AFL group (week 2). No significant differences were seen between any other groups, specifically between the Combo-DDS group and the AFL-DDS group.

Intraocular Pressure

Pre-injection IOP measurements averaged 19.9 ± 0.8 millimeters of mercury (mm Hg), which is within the normal range of IOP for rats (15–25 mm Hg).⁴³ Increases in IOP were seen for all groups immediately after injection (P < 0.05), however, measurements returned to baseline levels by week 2. Starting at week 2, all treatment groups maintained IOP levels similar within normal range for rats for the remainder of the study. Mean values for each time point are shown in Table 2.

Table 2.

View Table

Summary of IOP Measurements for CNV Study

Histology

Histopathology confirmed the formation of CNV lesions characterized by the presence of full-thickness retinal atrophy consistent with laser-induced models. Representative H&E slides can be seen in Figure 4. Relative thickness of the retina was determined as 4.26 ± 1.31, 2.9 ± 0.20, 1.86 ± 0.47, 1.87 ± 0.20, 1.26 ± 0.11, and 1.19 ± 0.06 for the No Treatment group, the Blank-DDS group, the Bolus AFL group, the AFL-DDS group, the DEX-DDS group, and the Combo-DDS group, respectively. These measurements were consistent with the initial subjective histomorphologic evaluation, which determined that the retinal lesions were more severe in the No Treatment group, moderately severe in the AFL-DDS group, and less severe in the Combo-DDS group. Statistically significant differences were found in the AFL-DDS group and the Combo-DDS group compared to the No Treatment group (P < 0.05), however, there were no differences between the AFL-DDS group and the Combo-DDS group. Although the Blank-DDS group had larger relative thickness than all treatment groups (Bolus AFL, AFL-DDS, DEX-DDS, and Combo-DDS), none of the differences were significant. There was no DDS material found during evaluation, indicating that DDS had degraded within the 22-week study. However, mild posterior chamber granulomatous reactions were found in 2 of the 16 eyes treated with DDS.

Figure 4.

View Original Download Slide

Histological evaluation of CNV lesions. (a) Representative histology images of CNV lesions from the No Treatment group (black), the Blank-DDS group (blue), the Bolus AFL group (pink), the AFL-DDS group (green), the DEX-DDS group (dark purple), and the Combo-DDS group (light purple). (b) Representative images for No Treatment. (c) Representative image for Combo-DDS. Statistically significant differences between groups were determined using Holm-Sidak multiple comparisons tests after repeated-measures 1-way ANOVA. Statistical differences of P <0.05 is represented by *.

Electroretinogram

ERG responses before and after Combo-DDS injection were used to determine any changes in outer and inner retinal function from a-wave and b-wave amplitudes, respectively. Maximum a-wave responses can be seen in Figure 5a. While a decreasing trend was observed throughout the study, the only significant decrease in maximum a-wave amplitude was at week 17 (32.9%; P = 0.0093). Two of the three rodents were exposed to light from a fire alarm system prior to ERG being conducted at week 17. Therefore, the decreases at week 17 may be due to disruption in dark adaption of the rodents. Figure 5b shows the b-wave intensity response function using Naka-Rushton analysis. No significant differences were observed during the study compared to control (baseline) data. Similarly, for maximum b-wave responses, the only significant difference was observed at week 17 (42.6%; P = 0.0134). A-wave and b-wave maximum amplitudes and sensitivities are summarized in Table 3.

Figure 5.

View Original Download Slide

ERG responses throughout the 6-month study for Combo-DDS. (a) Maximum a-wave values were responses to single light flashes of 311 sc·cd·s·m-2. (b) Average intensity-response function for a-wave with flash intensities ranging from 0.311 to 311 sc·cd·s·m². (c) Maximum b-wave values were responses to single light flashes between 0.000311 and 0.311 sc·cd·s·m-2. (d) Average intensity-response function for b-wave with flash intensities ranging from 0.000311 to 0.311 sc·cd·s·m². Solid lines represent Naka-Rushton analysis for (b) and (d). Error bars represent standard error (n = 6 eyes). Statistically significant difference of P < 0.05 represented by * and of P < 0.01 represented by **.

Table 3.

View Table

Summary of ERG Responses for Combo-DDS

Retinal Thickness

SD-OCT images were used to measure total retinal thickness away from DDS (control) and near DDS, as shown in Table 4. No significant differences were observed at any time point (P > 0.25).

Table 4.

View Table

Total Retinal Thickness Measurements for Combo-DDS

Discussion

This study demonstrated that our Combo-DDS has improved treatment efficacy compared with bimonthly aflibercept bolus injections in a laser-induced CNV rodent model evaluated for 6 months. The Combo-DDS also showed similar treatment efficacy to AFL-DDS. Additionally, Combo-DDS showed in vivo safety, tolerability, and biocompatibility when injected into the eyes of Long-Evans rats.

A laser-induced CNV rodent model was used to evaluate treatment efficacy due to its consistent CNV formation of about 60% to 80%, relatively low costs and maintenance, and the ability to be evaluated longitudinally over 6 months through FA and SD-OCT.⁴⁴ Previous studies in rodents have shown that CNV lesions require about 2 weeks to fully develop and are maintained for up to 6 months.⁴⁵^,⁴⁶ There is a great amount of variability in the natural regression of CNV lesions in regard to the time regression to begin⁴⁷; despite this, the No Treatment group maintained CNV lesions over the 6-month study. Similarly, our previous studies have shown consistency with non-treated CNVs without any signs of regression for at least 12 weeks after photocoagulation.²⁵^,³⁹ This allows for the conclusion that the extended reduction in CNV lesion size is indicative of the long-term treatment efficacy of AFL-DDS and Combo-DDS. Future studies should be utilized with NHPs for further evaluation of the DDS.

The MOT technique previously developed by our laboratory was used to quantify CNV lesions throughout the study. Previous studies have shown that values for the CNV area using MOT were consistent with standard histopathology measurements.³⁵ An advantage of MOT analysis is that it allows the evaluation of the same lesions over time. Histological evaluation performed in this study at the endpoint agreed with MOT analysis, finding AFL-DDS and Combo-DDS to have greater overall regression of lesions compared with all other treatment groups. MOT quantification found that average lesion size was smaller for the AFL-DDS group and the Combo-DDS group than all other groups at each time point, however, statistically significant differences were not always seen. Nevertheless, the Combo-DDS group showed statistically significant differences from the No Treatment group (weeks 0, 2, 14, and 22), the Blank-DDS group (weeks 6, 18, and 22), and the DEX-DDS group (week 14). It is important to note that comparison between the Combo-DDS group and the No Treatment group at week 18 gave P = 0.0689, which indicates a trend in reduction in lesion size of the Combo-DDS group compared to the No Treatment group at this time point.

Although the Combo-DDS group showed overall treatment efficacy, it did not show significant improvements compared to the AFL-DDS group. The DEX-DDS group, which has the same drug load for DEX as the Combo-DDS group, showed regression in overall lesion size at each time point, however, no significant differences were found when compared against the No Treatment and Blank-DDS groups. This indicates that a higher overall drug load for DEX may need to be delivered for Combo-DDS to show improved treatment efficacy compared to monotherapy AFL-DDS.

For ERG analysis, there were statistically significant decreases in maximum a- and b-wave only at week 17. After the decreases at week 17, maximum responses for both a- and b-wave increased for week 22 back to baseline values. Previous studies have shown that there is a 30% to 55% decrease in maximum a-wave and approximately 48% decrease in maximum b-wave responses between light- and dark-adapted ERGs.⁴⁸^,⁴⁹ Therefore, the changes at week 17 for maximum a-wave (32.9%) and maximum b-wave (42.6%) can be attributed to the known variance between dark- and light-adapted ERG and are unlikely to be physiologically significant. All other time points had similar values to baseline (pre-injection) over the 6-month study. This suggests that Combo-DDS does not cause any physiological changes to either the outer or inner retina. Furthermore, no signs of ocular toxicity were observed based on retinal thickness measurements.

Histology found that 2 of the 16 eyes treated with DDS had mild posterior chamber granulomatous reactions. In one of the eyes, the granulomatous reaction was caused by a focal break in the posterior lens capsule interpreted that likely happened during the intraocular injection procedure and was not related to the DDS treatment. The other eye had a less severe (minimal) granulomatous reaction around the DDS material present in the posterior chamber. No previous studies using PEG-PLLA-DA/NIPAAm hydrogel DDS had any similar reactions, and a definitive cause could not be defined. Endotoxins are known to cause uveitis and toxic anterior segment syndrome.⁵⁰^–⁵⁴ Therefore, all materials used in DDS fabrication were tested for possible contamination, and high levels of endotoxins were found in the water. It is unclear if DDS contamination caused the posterior chamber granulomatous reactions, however, several measures will be taken for future experiments to minimize any possible contamination: (1) sterile water will be used for fabrication of DDS and (2) all particles and DDS will undergo endotoxin testing prior to use. Future histological analysis can give further insight if granulomatous reactions were due to DDS or contamination.

One main limitation of this study is that the laser-induced CNV model in rodents is not representative of patients with non-responsive wet AMD. The rodent laser-induced model of CNV is the current standard animal model of CNV for treatment efficacy experiments, and, therefore, it is not expected to have any subjects that would not respond to anti-VEGF treatment.⁵⁵ However, NHP models could be used to gain more insight to the clinical relevance of the system due to their physiologic similarities and close phylogenetic relationships to humans.⁵⁵^,⁵⁶ Determination of in vivo degradation and pharmacokinetics in NHP models would then allow for further optimization of the DDS.

Last, the robustness and versatility of this DDS to simultaneously deliver combination therapy has been shown in this study. Although significant improvement in treatment efficacy in the CNV model was not seen, the Combo-DDS treatment has the potential to positively impact non-responsive patients. The Combo-DDS provides benefit by sustaining therapeutic drug levels without high peak drug concentrations associated with bolus dosing.⁵⁷^,⁵⁸ It may also improve patient adherence to therapy by reducing treatment burden and the need for separate treatment regimens with repeated intravitreal injections.⁵⁷^–⁶⁰ Additionally, this DDS can be adapted to incorporate other drugs,²⁴^,⁶¹ allowing for the development of other combination therapies, whether that is with different anti-VEGF agents, like faricimab and brolucizumab, or tyrosine kinase inhibitors (TKIs).

This study demonstrated that Combo-DDS maintained treatment efficacy in a laser-induced CNV rodent model for 6 months. Although statistically significant differences were not always found, Combo-DDS showed larger decrease in CNV lesion compared to No Treatment, Blank-DDS, Bolus AFL, and DEX-DDS. NHPs could be used to conduct dose ranging pharmacokinetic and pharmacodynamic studies for this combination DDS. This would allow for further optimization of each drug load amount and injection frequency.

Acknowledgments

Supported by the NIH under Grant EY029298.

Disclosure: K.M. Rudeen, AbbVie (E); C.M. Maloney, None; K.L. Lydon, None; L.B.C. Teixeria, None; W.F. Mieler, None; J.J. Kang-Mieler, microsphere-thermoresponsive drug delivery system (P)

References

Bourne RRA, Jonas JB, Bron AM, et al. Prevalence and causes of vision loss in high-income countries and in Eastern and Central Europe: 1990–2010. Br J Ophthalmol. 2014; 98(5): 629–638. [CrossRef] [PubMed]

Wong WL, Su X, Li X, et al. Global prevalence of age-related macular degeneration and disease burden projection for 2020 and 2040: a systematic review and meta-analysis. Lancet Glob Health. 2014; 2(2): e106–e116. [CrossRef] [PubMed]

Tan PL, Bowes Rickman C, Katsanis N. AMD and the alternative complement pathway: genetics and functional implications. Hum Genomics. 2016; 10(1): 23. [CrossRef] [PubMed]

Campochiaro PA, Soloway P, Ryan SJ, Miller JW. The pathogenesis of choroidal neovascularization in patients with age-related macular degeneration. Mol Vis. 1999; 5(11): 34. [PubMed]

Donoso LA, Kim D, Frost A, Callahan A, Hageman G. The role of inflammation in the pathogenesis of age-related macular degeneration. Surv Ophthalmol. 2006; 51(2): 137–152. [CrossRef] [PubMed]

Mathews MK, Merges C, McLeod DS, Lutty GA. Vascular endothelial growth factor and vascular permeability changes in human diabetic retinopathy. Invest Ophthalmol Vis Sci. 1997; 38(13): 2729–2741. [PubMed]

Aiello LP, Avery RL, Arrigg PG, et al. Vascular endothelial growth factor in ocular fluid of patients with diabetic retinopathy and other retinal disorders. N Engl J Med. 1994; 331(22): 1480–1487. [CrossRef] [PubMed]

Tolentino MJ, Miller JW, Gragoudas ES, Chatzistefanou K, Ferrara N, Adamis AP. Vascular endothelial growth factor is sufficient to produce iris neovascularization and neovascular glaucoma in a nonhuman primate. Arch Ophthalmol. 1996; 114(8): 964. [CrossRef] [PubMed]

Miller JW, Adamis AP, Shima DT, et al. Vascular endothelial growth factor/vascular permeability factor is temporally and spatially correlated with ocular angiogenesis in a primate model. Am J Pathol. 1994; 145(2): 574–584. [PubMed]

10.

Yi X, Ogata N, Komada M, et al. Vascular endothelial growth factor expression in choroidal neovascularization in rats. Graefes Arch Clin Exp Ophthalmol. 1997; 235(5): 313–319. [CrossRef] [PubMed]

11.

Eylea (aflibercept) injection for intravitreal use [package insert on the Internet]. Tarrytown, NY: Regeneron Pharmaceuticals, 2011 [revised 2019 May; cited 2020 Jan 9]. Available from: https://www.accessdata.fda.gov/drugsatfda_docs/label/2019/125387s061lbl.pdf.

12.

Oca AI, Pérez-Sala A, Pariente A, et al. Predictive biomarkers of age-related macular degeneration response to anti-VEGF treatment. J Pers Med. 2021; 11: 1329. [CrossRef] [PubMed]

13.

Wallsh JO, Gallemore RP. Anti-VEGF-resistant retinal diseases: a review of the latest treatment options. Cells. 2021; 10: 1049. [CrossRef] [PubMed]

14.

Yang S, Zhao J, Sun X. Resistance to anti-VEGF therapy in neovascular age-related macular degeneration: a comprehensive review. Drug Des Devel Ther. 2016; 10: 1857–1867. [CrossRef] [PubMed]

15.

Al-Khersan H, Hussain RM, Ciulla TA, Dugel PU. Innovative therapies for neovascular age-related macular degeneration. Expert Opin Pharmacother. 2019; 20(15): 1879–1891. [CrossRef] [PubMed]

16.

Forte R, Bonavolontà P, Benayoun Y, Adenis JP, Robert P-Y. Intravitreal ranibizumab and bevacizumab in combination with full-fluence verteporfin therapy and dexamethasone for exudative age-related macular degeneration. Ophthalmic Res. 2011; 45(3): 129–134. [CrossRef] [PubMed]

17.

Kaya C, Zandi S, Pfister I, Gerhardt C, Garweg JG. Adding a corticosteroid or switching to another anti-VEGF in insufficiently responsive wet age-related macular degeneration. Clin Ophthalmol. 2019; 13: 2403–2409. [CrossRef] [PubMed]

18.

Couch SM,, Bakri SJ. Review of combination therapies for neovascular age-related macular degeneration. Semin Ophthalmol. 2011; 26(3): 114–120. [CrossRef] [PubMed]

19.

Smith AG,, Kaiser PK. Emerging treatments for wet age-related macular degeneration. Expert Opin Emerg Drugs. 2014; 19(1): 157–164. [CrossRef] [PubMed]

20.

Villegas VM, Aranguren LA, Kovach JL, Schwartz SG, Flynn HW, Jr. Current advances in the treatment of neovascular age-related macular degeneration. Expert Opin Drug Deliv. 2017; 14(2): 273–282. [CrossRef] [PubMed]

21.

Conti SM,, Kertes PJ. The use of intravitreal corticosteroids, evidence-based and otherwise. Curr Opin Ophthalmol. 2006; 17(3): 235–244. [CrossRef] [PubMed]

22.

Ranchod TM, Ray SK, Daniels SA, Leong CJ, Tin TD, Verne AZ. LuceDex: a prospective study comparing ranibizumab plus dexamethasone combination therapy versus ranibizumab monotherapy for neovascular age-related macular degeneration. Retina. 2013; 33(8): 1600–1604. [CrossRef] [PubMed]

23.

Liu W, Lee B-S, Mieler WF, Kang-Mieler JJ. Biodegradable microsphere-hydrogel ocular drug delivery system for controlled and extended release of bioactive aflibercept in vitro. Curr Eye Res. 2018; 44(3): 264–274. [CrossRef] [PubMed]

24.

Liu W, Borrell MA, Venerus DC, Mieler WF, Kang-Mieler JJ. Characterization of biodegradable microsphere-hydrogel ocular drug delivery system for controlled and extended release of ranibizumab. Transl Vis Sci Technol. 2019; 8(1): 12. [CrossRef] [PubMed]

25.

Liu W, Tawakol AP, Rudeen KM, Mieler WF, Kang-Mieler JJ. Treatment efficacy and biocompatibility of a biodegradable aflibercept-loaded microsphere-hydrogel drug delivery system. Transl Vis Sci Technol. 2020; 9(11): 13. [CrossRef] [PubMed]

26.

Kim S, Kang-Mieler JJ, Liu W, et al. Safety and biocompatibility of aflibercept-loaded microsphere thermo-responsive hydrogel drug delivery system in a nonhuman primate model. Transl Vis Sci Technol. 2020; 9(3): 30. [CrossRef]

27.

Osswald CR, Kang-Mieler JJ. Controlled and extended release of a model protein from a microsphere-hydrogel drug delivery system. Ann Biomed Eng. 2015; 43(11): 2609–2617. [CrossRef] [PubMed]

28.

Patel NB, Luo X, Wheat JL, Harwerth RS. Retinal nerve fiber layer assessment: area versus thickness measurements from elliptical scans centered on the optic nerve. Invest Opthalmol Vis Sci. 2011; 52(5): 2477–2489. [CrossRef]

29.

Guthrie MJ, Osswald CR, Valio NL, Mieler WF, Kang-Mieler JJ. Objective area measurement technique for choroidal neovascularization from fluorescein angiography. Microvasc Res. 2014; 91: 1–7. [CrossRef] [PubMed]

30.

Rudeen KM, Liu W, Mieler WF, Kang-Mieler JJ. Simultaneous release of aflibercept and dexamethasone from an ocular drug delivery system. Curr Eye Res. 2022; 47(7): 1034–1042. [CrossRef] [PubMed]

31.

Gómez-Gaete C, Tsapis N, Besnard M, Bochot A, Fattal E. Encapsulation of dexamethasone into biodegradable polymeric nanoparticles. Int J Pharm. 2007; 331(2): 153–159. [CrossRef] [PubMed]

32.

Herrero-Vanrell R, Bravo-Osuna I, Andrés-Guerrero V, Vicario-de-la-Torre M, Molina-Martínez IT. The potential of using biodegradable microspheres in retinal diseases and other intraocular pathologies. Prog Retin Eye Res. 2014; 42: 27–43. [CrossRef] [PubMed]

33.

Drapala PW, Brey EM, Mieler WF, Venerus DC, Kang Derwent JJ, Pérez-Luna VH. Role of thermo-responsiveness and poly(ethylene glycol) diacrylate cross-link density on protein release from poly(N-isopropylacrylamide) hydrogels. J Biomater Sci Polym Ed. 2011; 22(1): 59–75. [PubMed]

34.

Grossniklaus HE, Kang SJ, Berglin L. Animal models of choroidal and retinal neovascularization. Prog Retin Eye Res. 2010; 29(6): 500–519. [CrossRef] [PubMed]

35.

Kuroki AM, Bhutto IA, Kitaoka T, Amemiya T. Natural course of experimental choroidal neovascularization: three-dimensional study with corrosion cast and scanning electron microscope. Ophthalmic Res. 2002; 34(4): 200–205. [CrossRef] [PubMed]

36.

Ming Y, Algvere PV, Odergren A, et al. Subthreshold transpupillary thermotherapy reduces experimental choroidal neovascularization in the mouse without collateral damage to the neural retina. Invest Opthalmol Vis Sci. 2004; 45(6): 1969–1974. [CrossRef]

37.

Turturro SB, Guthrie MJ, Appel AA, et al. The effects of cross-linked thermo-responsive PNIPAAm-based hydrogel injection on retinal function. Biomaterials. 2011; 32(14): 3620–3626. [CrossRef] [PubMed]

38.

Lima LH, Morales Y, Cabral T. Ocular biocompatibility of poly-N-isopropylacrylamide (pNIPAM). J Ophthalmol. 2016; 2016: 5356371. [CrossRef] [PubMed]

39.

Osswald CR, Guthrie MJ, Avila A, Valio JA, Jr, Mieler WF, Kang-Mieler JJ. In vivo efficacy of an injectable microsphere-hydrogel ocular drug delivery system. Curr Eye Res. 2017; 42(9): 1293–1301. [CrossRef] [PubMed]

40.

Kang Derwent JJ, Linsenmeier RA. Intraretinal analysis of the a-wave of the electroretinogram (ERG) in dark-adapted intact cat retina. Vis Neurosci. 2001; 18(3): 353–363. [CrossRef] [PubMed]

41.

Lamb TD, Pugh EN. A quantitative account of the activation steps involved in phototransduction in amphibian photoreceptors. J Physiol. 1992; 449(1): 719–758. [CrossRef] [PubMed]

42.

Sadaka A, Giuliari GP. Proliferative vitreoretinopathy: current and emerging treatments. Clin Ophthalmol. 2012; 6: 1325–1333. [PubMed]

43.

Pang IH, Wang WH, Clark AF. Acute effects of glaucoma medications on rat intraocular pressure. Exp Eye Res. 2005; 80(2): 207–214. [CrossRef] [PubMed]

44.

Pennesi ME, Neuringer M, Courtney RJ. Animal models of age related macular degeneration. Mol Aspects Med. 2012; 33(4): 487–509. [CrossRef] [PubMed]

45.

Grossniklaus HE, Kang SJ, Berglin L. Animal models of choroidal and retinal neovascularization. Prog Retin Eye Res. 2010; 29(6): 500–519. [CrossRef] [PubMed]

46.

47.

Pollack A, Korte GE. Repair of retinal pigment epithelium and choriocapillaries after laser photocoagulation: correlations between scanning electron, transmission electron and light microscopy. Ophthalmic Res. 1997; 29(6): 393–404. [CrossRef] [PubMed]

48.

Lachapelle P. Analysis of the photopic electroretinogram recorded before and after dark adaptation. Can J Ophthalmol. 1987; 22(7): 354–361. [PubMed]

49.

Thomas MM, Light adaptation and dark adaptation of human rod photoreceptors measured from the a-wave of the electroretinogram. J Physiol. 1999; 518(Pt 2): 479–496. [PubMed]

50.

Bantseev V, Miller PE, Bentley E, et al. Determination of a no-observable effect level for endotoxin following a single intravitreal administration to Dutch Belted rabbits. Invest Ophthalmol Vis Sci. 2017; 58(3): 1545–1552. [CrossRef] [PubMed]

51.

Forrester JV, Klaska IP, Yu T, Kuffova L. Uveitis in mouse and man. Int Rev Immunol. 2012; 32(1): 76–96. [CrossRef]

52.

Csukas S, Paterson CA, Brown K, Bhattacherjee P. Time course of rabbit ocular inflammatory response and mediator release after intravitreal endotoxin. Invest Ophthalmol Vis Sci. 1990; 31(2): 382–387. [PubMed]

53.

Gilger BC, Abarca EM, Salmon JH, Patel S. Treatment of acute posterior uveitis in a porcine model by injection of triamcinolone acetonide into the suprachoroidal space using microneedles. Invest Ophthalmol Vis Sci. 2013; 54(4): 2483–2492. [CrossRef] [PubMed]

54.

Denlinger JL, El-Mofty AA, Balazs EA. Replacement of the liquid vitreous with sodium hyaluronate in monkeys. II. Long-term evaluation. Exp Eye Res. 1980; 31(1): 101–117. [CrossRef] [PubMed]

55.

Grossniklaus HE, Kang SJ, Berglin L. Animal models of choroidal and retinal neovascularization. Prog Retin Eye Res. 2010; 29(6): 500–519. [CrossRef] [PubMed]

56.

Chen S, Popp NA, Chan CC. Animal models of age-related macular degeneration and their translatability into the clinic. Expert Rev Ophthalmol. 2014; 9(4): 285–295. [CrossRef] [PubMed]

57.

Lee SS, Hughes P, Ross AD, Robinson MR. Biodegradable implants for sustained drug release in the eye. Pharm Res. 2010; 27: 2043–2053. [CrossRef] [PubMed]

58.

Yasukawa T, Ogura Y, Kimura H, SakuraiTabataE, Drug delivery from ocular implants. Expert Opin Drug Deliv. 2006; 3(2): 261–273. [CrossRef] [PubMed]

59.

Bourges J, Bloquel C, Thomas A, et al. Intra-ocular implants for extended drug delivery: therapeutic applications. Adv Drug Deliv Rev. 2006; 58: 1182–1202. [CrossRef] [PubMed]

60.

Wang J, Jiang A, Joshi M, Christoforidis J. Drug delivery implants in the treatment of vitreous inflammation. Mediators Inflamm. 2013; 2013:780634. [PubMed]

61.

Yuan B, Rudeen K, Li J, et al. Biodegradable microspheres and hydrogel drug delivery system of tumor necrosis factor (TNF) inhibitor and growth differentiation factor 5 (GDF5) reduces disk inflammation in the rabbit model. Spine. 2023; 48(15): E257–E265. [CrossRef] [PubMed]