Dynamic versus Static In Vitro Drug Release Testing of Subcutaneous Implants with Distinct Microstructures

Implantable long-acting drug delivery systems, such as subcutaneous implants, provide controlled, sustained release of an active ingredient. Although drug release kinetics are critical in the development of such systems, current in vitro release testing methods often fail to account for key physiological parameters such as pH, buffer capacity, flow conditions, as well as tissue/extracellular matrix (ECM) firmness and proteins. This study investigates the long-term in vitro release behavior of two biodegradable dexamethasone-loaded implants with different internal structures: dense implants produced via hot-melt extrusion (HME, 100% infill) and porous implants fabricated via fused filament fabrication (FFF, 25% infill). Phosphate buffers (10 mM and 100 mM), simulated body fluid, and simulated subcutaneous interstitial fluid were evaluated as potential release media. Among the tested buffers, the 100 mM phosphate buffer showed the highest long-term stability, maintaining a consistent buffer capacity with minimal fluctuations in pH and osmolarity and no precipitation at elevated temperatures. Protein-containing media, in general, showed precipitation and degradation and were therefore excluded from further studies. The effect of interstitial fluid flow was evaluated using the USP 4 apparatus, and release was studied over a period of 8 weeks. It was found thatcompared to static conditionsdrug release from porous FFF implants significantly increased from 23.71 ± 1.07% to 25.99 ± 0.41%. In contrast, dense HME implants exhibited diffusion-dominated release that was unaffected by flow. The underlying release mechanism, as determined by Korsmeyer-Peppas modeling, was dominated by diffusion for both implants, consistent with the behavior observed under static release conditions. The influence of tissue/ECM firmness was investigated in a gel-based setup using a 0.5% agarose gel, which allows diffusion of small molecules, mimics the bulk viscoelastic behavior of soft connective tissues/ECM, and shows comparable pore sizes. As expected, the overall drug release and underlying mechanism changed for both implant types, resulting in a lower absolute drug release and elevated n-values. Our findings provide fundamental insights into the interplay between relevant parameters, in particular flow and tissue/ECM firmness, and internal implant structures on the in vitro drug release, offering a framework to improve long-term testing of subcutaneous implants.