(W1030-06-31) Small Scale, Rapid Prototyping of Hydralese™ (PGSU) (poly(glycerol sebacate) urethane) Microdevices for Controlled Drug Delivery in Ocular Implants

Purpose: We are developing long-acting implantable devices manufactured from Hydralese™ (PGSU) (poly(glycerol sebacate) urethane) for controlled drug release over a multi-month time period. Hydralese (PGSU) devices are currently fabricated by crosslinking drug-loaded poly(glycerol sebacate) (PGS) resin via polyol-isocyanate urethane chemistry using a laboratory scale dual barrel syringe (DBS). The DBS injects the drug-polyol blend and the isocyanate crosslinker through separate barrels into a static mixing tip and the blended material is flowed into a mold for ambient curing. Our validated DBS Hydralese (PGSU) extrusion method has been useful for rapid development of formulations and testing prototype geometries. However, it is limited in extrusion pressures with high viscosity drug/resin mixtures and inability to control the material temperature within. These limitations created challenges in producing high drug loading formulations at small device sizes. Ocular implants require high drug loading (≥50% API w/w) with device sizes at sub millimeter ranges. High pressure differentials into the small tube molds slowed down the reactive mixture flow rate in the upstream static mixing region causing the mixture to cure and not flow. Additionally, timeline requirements necessitated rapid formulation at various microdevice diameters and eliminated precision-machined stainless steel molds as an option due to long lead times and high capital cost commitments. To combat these production limitations, a low cost, rapid prototyping method has been developed to create high drug loading microdevices for long-acting drug delivery. These microdevices were developed for ocular indications and were dexamethasone-loaded intra-vitreal implants. The main process improvements used microtube parallelized bundles in combination with heating of the DBS cartridge. The microtube bundle emulates a multi-lumen extrusion to reduce differential pressures and increase flow rate, while the temperature control of the reactive materials reduces and maintains the viscosity of high drug loading resin blends over the course of the mold injection process. The aims of this study were: 1) to develop a prototyping method for casting 200µm and 450µm outer diameter x 20mm long microdevices at loadings up to 60% API w/w 2) to characterize the morphology, diameter, and loading of the loaded Hydralese (PGSU) microdevices.

Methods: PTFE microtubing was sourced at commercially available inner diameters of 0.008” (203µm) and 0.018” (457µm) to approximate customer target diameters of 200µm and 450µm respectively. Microtubes were cut to a 30mm length, and a predetermined number of microtubes were inserted into an equal length 0.09” ID PTFE heat shrink tubing. Bundles of parallel microtubes sheathed in heat shrink tubing were heated and allowed to shrink, securing the microtubing together. The bundle of microtubes was inserted through a Luer-lock dispensing tip The bundle was mated to a Luer-lock static mixing tip and allowed to fill from the dispense end of the static mixer. In order to reduce the formulated material viscosity, temperature control was added to the system with the combination of 54W heat trace wrapped in felt insulation. The felt insulation created a warming wrap to improve heating contact area with the DBS and dampen the intensity of the heat trace. The felt-covered heat trace wrap was combined with a thermocouple and thermostatic heating controller to provide temperature control of 35±2.5°C. 60% w/w dexamethasone PGSU microrods were formulated and fabricated by a microtube bundle molding process. The polyol (A-component) was comprised of pre-polymer PGS resin loaded with dexamethasone and a tin catalyst. The isocyanate (B-component) was comprised of pre-polymer PGS resin loaded with dexamethasone and mixed with hexamethylene diisocyanate (HDI) to create an overall 3.5:1 PGS:HDI w/w crosslink ratio. The component mixtures were heated to 35±2.5°C using the heating system. Using a pneumatic piston driver to maintain constant simultaneous pressure on the component chambers, materials were flowed through a static mixing tip into the microtube bundles. Formulated material was cured in the mold for 24 hours at 23°C before demolding and inspecting product microdevices. Resulting drug-loaded PGSU microrods were examined for shape and surface finish using scanning electron microscopy (SEM). Microrod outer diameter was measured by calibrated optical microscopy and compared against dimensional tolerances. The drug loading was characterized by solvent extraction in DMSO and quantification by HPLC.

Results: 60% w/w dexamethasone Hydralese (PGSU) microrods were successfully made using microtube bundles at 203µm and 457µm nominal outer diameter. The SEM images show the rods were cylindrical with a smooth finished exterior. Microdevice outer diameter, measured by calibrated optical microscopy, demonstrated dimensional tolerances within ± 10µm. Drug loading were measured by extraction and HPLC and found to be on target. Dexamethasone-loaded Hydralese (PGSU) microrods exhibited morphology, diameter, and loading consistent with industry expectations and tolerances.

Conclusion: The microtube bundle process is an effective prototyping method for high drug-loaded Hydralese (PGSU) formulations and corresponding microdevice fabrication. The microtube bundle process allows for increased production volumes while maintaining control over microdevice quality. This process can be adapted for scale-up to support production of drug-loaded PGSU intravitreal implants for ocular drug delivery.

Optical microscopy of 457µm nominal diameter Hydralese™ (PGSU) microrod with 60% w/w dexamethasone loading showing in tolerance (±10µm) diameter and smooth surface finish


Example of 457µm X 7count X 30mm long microtube bundle with cross section



Dual barrel syringe (DBS) setup with insulated 54W heat trace cable.