(M1430-11-76) Understanding the Impact of Accelerated Release Testing Conditions on Transport Properties and Release Mechanisms of Drug from Long-Acting Ethylene Vinyl Acetate Implants

Purpose: Ethylene vinyl acetate (EVA)-based long-acting reservoir implants such as Nexplanon etonogestrel implant consist of a dispersed drug-containing polymer core and an EVA polymer skin layer. Drug release can be controlled based on material transport properties and implant dimensions. The general release mechanism is based on diffusion, however since the skin layer does not cover the ends, the impact of drug dissolution on release through ends and skin are different. Such implants are designed to deliver drug over multiple years, necessitating accelerated release test methods to facilitate reasonable development times and product release testing. The purpose of this study is to elucidate the impact of accelerated conditions on drug release from dispersed-drug EVA-based implants.

Methods: EVA 15 (15% vinyl acetate) films (79 μm) were manufactured using a Randcastle single-screw extruder fitted with a sheet die and film puller. Etonogestrel solubility in EVA 15 was measured by saturation experiments. Diffusivity was determined using Permegear side-by-side diffusion cells. A film was clamped between donor and receptor cells containing drug suspension and blank media, respectively. To study the impact of solvent, different compositions of EtOH:H2O were used at 37°C. To study the impact of temperature, 0.00075% Tween 80 in H2O was used while temperature was varied. Drug content was assayed using HPLC. Films were stored at elevated temperatures for 12 hours or in 75:25 EtOH:H2O for 3 weeks, and crystal structure was assessed using differential scanning calorimetry. Implants were manufactured using a coextrusion process consisting of a single-screw extruder (Davis-Standard) and a twin-screw extruder (Leistritz) combined with a water bath, laser micrometer, and puller. For further detailed discussion on the implant manufacturing process and real-time release mechanisms please refer to the abstract titled “Manufacturing and elucidating the drug release mechanisms of long-acting EVA implants”. In brief, the core and the skin (70 μm) are made of EVA 28 and EVA 15, respectively. One implant is 4 cm long, 2.1 mm in diameter, and contains 38 mg drug. The core contained 30% drug with most present as solid particles and a small fraction as dissolved in the EVA matrix. Release was conducted using an incubator shaker at 37°C and 150 rpm. To analyze release through ends only (2 cm long) or skin only (2 cm long), implants were sealed with Loctite 4011 glue which is impermeable to etonogestrel. Real-time and accelerated release were conducted in 0.00075% (w/v) Tween 80 in H2O and 90:10 EtOH:H2O, respectively. The accelerated media was selected based Nexplanon test methods, a contraceptive implant with similar release behavior [1].

Results: Diffusivity was calculated from the release rate measured in diffusion experiments (Figure 1(a)) [1]. The permeability increase with % EtOH is primarily caused by increase in diffusivity due to change in free volume [2, 3]. The effect of temperature on release rate fits the Arrhenius equation (Figure 1(b)) [4]. DSC results indicate no impact of EtOH on crystal structure of the EVA polymer. The lower and higher temperature melting events correspond to melting of smaller, less perfect and larger, more perfect crystalline regions of the polymer, respectively. The less perfect crystalline regions form due to incorporation of amorphous regions [5]. Storage at elevated temperature resulted in shifting of the lower melting event to a higher temperature (Figure 2), indicating reorganization of the amorphous and crystalline regions. Since drug diffusion can only occur through the amorphous regions of EVA, size and distribution of crystalline EVA impact transport properties by changing diffusion path length [6, 7]. Figure 3 summarizes the real-time and accelerated release from skin only and ends only. Cumulative real-time release from skin was more than two times greater than from ends (Figure 3(a)), whereas cumulative accelerated release exhibited nearly identical behavior from skin and ends (Figure 3(b)). The release mechanism remains diffusion-based, but clearly the presence of organic solvent is accelerating release while changing the driving force for drug dissolution and diffusion in the polymer. Release from dispersed drug systems occurs by 1) drug dissolution into the polymer 2) diffusion to the matrix/media interface and 3) partition into the release media. The dissolution rate/diffusion rate ratio determines the rate-limiting step. We hypothesize that the ratio changes when switching from real-time to accelerated media which is reflected in the drastic change in curve shape (Figure 3(c) and 3(d)). The rate at which dissolved drug replenishes diffused drug determines the concentration gradient within the implant, which determines the correlation between release rate and time.

Conclusion: Drug release can be accelerated using organic solvent or elevated temperature. The relationship between permeability and temperature follows the Arrhenius equation, however elevated temperature causes changes in crystalline structure of EVA. Organic solvent increases permeability primarily by increasing diffusivity. Accelerated release based on organic solvent maintains a diffusion-based mechanism but changes the driving force for release from dispersed-drug implants by changing the dissolution rate/diffusion rate ratio.

References: 1. Lee, E., et al., Transport of steroids in poly(etherurethane) and poly(ethylene vinyl acetate) membranes. Journal of Membrane Science, 1985. 24: p. 125-143.
2. Chen, S., Diffusion of benzocaine in poly(ethylene-vinyl acetate) membranes: Effects of vehicle ethanol concentration and membrane vinyl acetate content. 1995.
3. Yasuda, H., C.E. Lamaze, and A. Peterlin, Diffusive and hydraulic permeabilities of water in water-swollen polymer membranes. Journal of Polymer Science Part A-2: Polymer Physics, 1971. 9(6): p. 1117-1131.
4. Kalachandra, S., et al., Drug release from cast films of ethylene vinyl acetate (EVA) copolymer: Stability of drugs by 1H NMR and solid state 13C CP/MAS NMR. Journal of Materials Science: Materials in Medicine, 2005. 16(7): p. 597-605.
5. Wang, K. and Q. Deng, The Thermal and Mechanical Properties of Poly(ethylene-co-vinyl acetate) Random Copolymers (PEVA) and its Covalently Crosslinked Analogues (cPEVA). Polymers (Basel), 2019. 11(6).
6. Chen, S., J. Zhang, and J. Su, Effect of damp-heat aging on the properties of ethylene-vinyl acetate copolymer and ethylene- acrylic acid copolymer blends. Journal of Applied Polymer Science, 2009. 114(5): p. 3110-3117.
7. Van Laarhoven, J.A.H., Physical-chemical aspects of a coaxial sustained release device based on poly-eva. 2005.

Acknowledgements: This work was supported by the Broad Agency Announcement (BAA) Contract # 75F40122C00019 from the U.S. Food and Drug Administration (FDA). The content reflects the views of the authors and should not be construed to present FDA’s views or policies. The authors wish to express their gratitude to Celanese for providing EVA for this study, and for their valuable insight related to EVA characterization and material properties.

Figure 1. Impact of (a) solvent and (b) temperature on transport properties of etonogestrel in EVA 15 as determined by side-by-side diffusion.

Figure 2. Impact of (a) solvent and (b) temperature on melting behavior of EVA 15 as determined by DSC. The lower and higher temperature melting peaks correlate to imperfect, smaller crystals and perfect, larger crystals, respectively.

Figure 3. Summary of implant release behavior (2 cm long). (a) Real-time cumulative release through skin only vs. ends only (b) accelerated cumulative release through skin only vs. ends only (c) real-time vs. accelerated release rate through skin only (d) real-time vs. accelerated release rate through ends only. Data represents the mean and standard deviation of n=3 replicates.