(M1030-07-43) Long-Term Release Kinetics and Characterizations of Episcleral Silicone Implants for Anterior Segment Ocular Delivery of Ketotifen Fumarate and Cyclosporine A

Purpose: Effective ocular drug delivery is hindered by anatomical barriers, especially in the anterior segment, where less than 5% is absorbed due to the precorneal barrier. Conventional treatments like eye drops face rapid clearance, low bioavailability, systemic effects, and poor compliance. Although therapies exist for conditions like dry eye, allergic conjunctivitis, glaucoma, and uveitis, many unmet clinical needs persist [1, 2]. To overcome these issues, non-biodegradable polymeric implants have been explored to enhance drug retention, provide sustained release, and improve treatment outcomes. Silicone matrix implants, in which drugs are uniformly embedded or surface-loaded, offer long-term release potential. This study focuses on characterizing the morphology and thermal behavior, validating HPLC methods, and evaluating the in vitro release of ketotifen fumarate (KF) and cyclosporine (CyA) from episcleral silicone matrix implants.

Methods: Medical-grade silicone elastomer-based episcleral implants [3] containing 10% and 30% of ketotifen fumarate (a weakly basic drug) and cyclosporine A (a neutral drug with weakly basic characteristics) were used to investigate how drug physicochemical properties influence release behavior from silicone matrices. Implant morphology and thermal characteristics were evaluated using scanning electron microscopy (SEM) and differential scanning calorimetry (DSC). Drug content was quantified using high-performance liquid chromatography (HPLC), with methods developed and validated for each drug, assessing linearity, accuracy, precision, and sensitivity, including limits of detection (LOD) and quantification (LOQ). In vitro drug release studies were conducted in phosphate-buffered saline (PBS) with ethanol (80:20, v/v) at 37 °C under sink conditions. The release medium was fully replaced at each sampling point, and drug release was monitored over 250 days. Collected samples were analyzed using validated HPLC methods, and the release profiles were modeled using zero-order, first-order, Higuchi, and Hixson–Crowell equations to evaluate release mechanisms and rate-controlling factors.

Results: The implants were approximately 1.8–2.0 cm in length and had a semicylindrical shape, rounded on one side and flat on the other, reflecting consistent size and geometry across formulations, as shown in Figure 1(a). Implants containing KF and CyA appeared pale or off-white, which intensified with higher drug loading. SEM micrographs in Figure 1(b) confirmed smooth external surfaces and uniform internal drug distribution without visible aggregates or phase separation. Cross-sectional images showed consistent dispersion across formulations, although slightly rougher textures appeared at higher drug concentrations. DSC thermograms revealed melting transitions for pure KF (~198.3 °C), which remained detectable in drug-loaded implants but with reduced intensity and minor shifts in temperature, indicating changes in thermal behavior upon incorporation, as shown in Figures 1(c) and 1(d). CyA showed no thermal transition in either form, consistent with its amorphous nature. Chromatographic methods were individually developed and validated for each drug, and the resulting chromatograms demonstrated well-resolved peaks with no interference, confirming high selectivity, as shown in Figure 2(a). Calibration curves exhibited excellent linearity, with R² = 0.9999 for KF and 0.9998 for CyA. Sensitivity analysis revealed that KF exhibited the lowest LOD/LOQ values (0.81/2.45 µg/mL), while CyA showed the highest (3.44/10.41 µg/mL), likely attributable to its hydrophobicity and larger molecular structure. All system suitability parameters met USP criteria. KF had optimal peak symmetry (tailing factor 0.92 ± 0.003) and the highest column efficiency (3080.06 ± 46.46 plates), and all RSDs were < 1%. The recovery percentages of ketotifen fumarate (KF) and cyclosporine A (CyA) extracted from the silicone matrix implants were 98.93 ± 1.33% and 99.81 ± 1.88%, respectively, as shown in Figure 2(b). Furthermore, the in vitro release of KF from the implants over 250 days demonstrated sustained drug release at a higher percentage for KF10 (34.14%) compared to KF30 (15.59%). Likewise, the in vitro release of CyA from the implants showed a sustained drug release at a higher percentage for CyA10 (61.84%) compared to CyA30 (32.12%) (Figure 1a). It appeared that KF had a strong affinity for the silicone matrix compared to CyA. A consistent pattern was observed in the cumulative amount released (Figure 1b), where higher drug loading resulted in greater total drug release for both drugs, with CyA30 and KF30 releasing 2717.76 µg and 1409.82 µg, respectively. Kinetic modeling (Figures 1c–d) showed that all formulations fit best with the Higuchi and first-order models (R² > 0.99), indicating that drug release was primarily governed by diffusion and concentration-dependent mechanisms. Zero-order and Hixson–Crowell models showed moderate correlation, suggesting limited involvement of constant-rate release or surface area-driven erosion. Overall, these results confirm that the silicone matrix system enables long-acting, diffusion-controlled drug release influenced by drug properties and loading concentrations.

Conclusion: This study successfully demonstrated that drug-loaded episcleral silicone matrix implants provide sustained in vitro release of ketotifen fumarate and cyclosporine over 250 days, with release behavior influenced by drug properties and loading concentration. The implants exhibited uniform morphology, stable thermal characteristics, and excellent analytical performance, supporting their potential as tunable platforms for anterior segment ocular drug delivery.

References: 1. Patel A, Cholkar K, Agrahari V, Mitra AK. Ocular drug delivery systems: An overview. World journal of pharmacology. 2013 Jun 9;2(2). doi: 10.5497/wjp.v2.i2.47.
2. Morrison PW, Khutoryanskiy VV. Advances in Ophthalmic Drug Delivery. Therapeutic Delivery. 2014;5(12):1297-315. doi: 10.4155/tde.14.75.
3. Kim H, Csaky KG, Gilger BC, Dunn JP, Lee SS, Tremblay M, et al. Preclinical Evaluation of a Novel Episcleral Cyclosporine Implant for Ocular Graft-Versus-Host Disease. Investigative Opthalmology & Visual Science. 2005;46(2):655. doi: 10.1167/iovs.04-1076.

Acknowledgements: This work was partly funded by the National Institutes of Health (NIH), National Eye Institute (NEI), under Grant Number 1R15EY035057-01.


Figure 1. Characterization of drug-loaded episcleral silicone matrix implants: (a) Visual appearance of the episcleral silicone matrix implants containing 10% and 30% (w/w) of ketotifen fumarate and cyclosporine A. (b) SEM images showing representative surface and cross-sectional morphologies of ketotifen fumarate implants at 10% and 30% drug loadings. (c) DSC thermograms of pure drugs and drug-loaded implants, indicating melting transitions and crystallinity changes upon incorporation into the polymer matrix. (d) Table summarizing thermal properties and percent crystallinity of pure drugs and drug-loaded implants.


Figure 2. HPLC method development and validation for quantifying the release of ketotifen fumarate and cyclosporine A from episcleral silicone matrix implants: (a). HPLC chromatograms. (b). Parameters of HPLC method validation.


Figure 3. Long-term in vitro drug release and kinetic modeling of ketotifen fumarate (KF) and cyclosporine A (CyA) from episcleral silicone implants with 10% and 30% drug loading over 250 days: (a). % Cumulative percentage of drug released. (b) Cumulative amount of drug released (µg). (c) Kinetic parameters from five models. (d) Model-fitting plots for each formulation group (KF and CyA), illustrating release behavior across all kinetic models.