An aqueous continuous phase containing 2.0% w/v poly(vinyl alcohol) 88% hydrolyzed and 0.9% w/v sodium chloride is equilibrated to 18–22 °C. A dispersed phase is prepared by dissolving RESOMER RG 858 S at 14% w/w in dichloromethane at 20 °C until the polymer is fully solvated and the solution passes a 0.22 µm PTFE filter. The inherent viscosity specification of 1.3–1.7 dL/g, measured as 0.1% in chloroform at 25 °C, governs the maximum polymer concentration that remains dispersible under rotor-stator shear. A primary water-in-oil emulsion is formed by adding an aqueous peptide solution equilibrated to 4 °C at a water-to-oil ratio of 1:4 and dispersing with an IKA T25 digital Ultra-Turrax at 24,000 rpm for 2 min. The primary emulsion is then transferred into the continuous aqueous phase and processed with a Silverson L5M-A rotor-stator mixer at 5,000 rpm for 10 min at 15 °C to generate a double-emulsion template. Solvent extraction proceeds by controlled linear addition of 1.0 L water at 0.5 mL/min, followed by reduced-pressure evaporation at 300 mbar for 3 h. The hardened microspheres are wet-sieved through 25 µm and 125 µm stainless steel screens, collected by filtration, and lyophilized in a SP Scientific Lyostar 3 freeze dryer with a shelf ramp from -45 °C to 20 °C at 0.05 mbar for 36 h. Mannitol is added at 5% w/w in the final resuspension vehicle rather than in the freeze-drying matrix to maintain cake structure and improve powder flow. The terminal product is a sterile injectable suspension after reconstitution in a vehicle containing 0.5% w/w sodium carboxymethylcellulose, 0.1% w/w polysorbate 80, and 4.5% w/w mannitol. Release testing uses USP Apparatus 4 in pH 7.4 phosphate-buffered saline at 37 °C, with sampling intervals aligned to the intended 1-month dosing interval. Compliance prerequisites include USP <71>, USP <85>, USP <788>, USP <790>, ICH Q3C residual dichloromethane below 600 ppm, USP <467> method suitability for solvent-specific headspace response, and ISO 10993-6 implantation evaluation for musculoskeletal tissue where the microsphere depot is administered subcutaneously or intramuscularly. Residual poly(vinyl alcohol) is quantified by a validated colorimetric method and controlled below 0.5% w/w of the lyophilized mass because higher residual emulsifier alters the initial drug release slope. Batch-to-batch particle size variability on production-scale colloid mills originates from non-constant temperature in the secondary emulsion vessel; jacketed stainless steel vessels with ±1 °C control are specified to maintain D50 between 30 µm and 80 µm. The process boundary is defined by the ratio of dispersed phase viscosity to continuous phase shear stress; exceeding polymer concentration above 18% w/w in dichloromethane elevates the dispersed phase viscosity beyond stable droplet breakup under standard rotor-stator conditions and shifts the product toward agglomerates with residual solvent above the ICH Q3C limit.

Residual solvent limits applicable to the solvent-based processes above are summarized in the following matrix.

When High Melt Viscosity Restricts Die Pressure in Implant Extrusion

Before melt compounding, RESOMER RG 858 S is vacuum-dried at 40 °C for 24 h until Karl Fischer moisture is below 0.1% w/w. A powder blend containing 82% w/w polymer, 14% w/w buprenorphine hydrochloride, and 4% w/w polyethylene glycol 3350 is premixed in a tumble blender for 20 min. The formulation is fed into a Leistritz ZSE 18 HP-P twin-screw extruder with a 25:1 L/D ratio and barrel zones set at 85 °C, 95 °C, 100 °C, and 105 °C. Screw speed is maintained at 30–60 rpm, and die pressure is recorded between 1.5 MPa and 4.0 MPa. Torque is monitored because the high inherent viscosity of the 85:15 lactide:glycolide grade creates a narrow processing window: below 85 °C melt fracture appears at the die, and above 115 °C chain scission accelerates and reduces implant mechanical integrity. The extrudate is pulled through a 1.5 mm single-strand die, cooled on a controlled air knife to 15 °C, and cut into 1.8 cm rods. A laser micrometer records extrudate diameter every 15 min to detect die swell drift. The terminal product is a single-use subcutaneous implant loaded at 14% drug substance. In-process tests include drug content uniformity per USP <905>, moisture per Karl Fischer, and residual monomers by HPLC with limits of ≤0.5% lactide and ≤0.5% glycolide. In vitro release uses USP Apparatus 4 in pH 7.4 buffer at 37 °C, with sinkers to prevent implant floating. Biocompatibility follows ISO 10993-1, ISO 10993-5, ISO 10993-10, and ISO 10993-6 for local subcutaneous tissue response. Sterility assurance requires USP <71>, and endotoxins are controlled by USP <85>. The main process limitation is moisture ingress; if the dried blend is held at ambient relative humidity above 60% for more than 30 min, the release profile shifts toward a lag phase followed by accelerated release because polymer hydrolysis occurs before implantation. A nitrogen-purged glove bag with ≤5% RH is required for blend transfer and hopper loading. Published data for gamma sterilization of this specific formulation is limited; terminal gamma irradiation at 25 kGy is expected to reduce the number-average molecular weight and alter the erosion time, so aseptic processing with terminal ethylene oxide validation per ISO 11135 is used for heat-sensitive drug loads.

Nanoparticulate systems are prepared by injecting a filtered acetone phase containing 10 mg/mL RESOMER RG 858 S and 1.5 mg/mL dexamethasone into a 0.25% w/v poloxamer 188 aqueous phase at 2 mL/min under magnetic stirring. The acetone-to-water volumetric ratio is 1:10, and the aqueous phase is maintained at 4 °C to limit acetone diffusion rate and reduce particle coalescence. The high molecular weight of RG 858 S requires an organic phase viscosity below 4 mPa·s to produce a narrow particle size distribution under low-energy nanoprecipitation. After solvent removal under reduced pressure at 200 mbar for 2 h, the suspension is concentrated by tangential-flow filtration using a 300 kDa regenerated cellulose membrane. Lyophilization proceeds with 5% w/w trehalose as cryoprotectant at -50 °C and 0.03 mbar for 48 h. The terminal product is a lyophilized nanoparticle cake for intravenous infusion after reconstitution with 0.9% w/v sodium chloride. Particle size is measured by dynamic light scattering after reconstitution; the targeted D50 is 140–220 nm and polydispersity index ≤0.15. Subvisible particulate counts follow USP <788>, and residual acetone is controlled below 5000 ppm under ICH Q3C. In vivo evaluation follows ISO 10993-6 for intravenous compatibility. Process control is limited by the sensitivity of nanoprecipitation to acetone phase concentration; increasing polymer concentration above 15 mg/mL rapidly raises the particle D50 beyond the intended biodistribution window, while decreasing below 5 mg/mL reduces entrapment efficiency and forces lyophilization cycle extension.

In Situ Phase-Inversion Depot Rheology and NMP Retention Limits

A formulation containing 42% w/w RESOMER RG 858 S, 7% w/w leuprolide acetate, 46% w/w N-methyl-2-pyrrolidone, and 5% w/w benzyl alcohol is prepared in a nitrogen-purged glass reactor at 25 °C until visual clarity is achieved under low-light inspection. The solution is filtered through a 0.45 µm hydrophilic PTFE membrane and filled into 1.0 mL cyclic olefin copolymer syringes. Rotational rheometry on an Anton Paar MCR 302 at 25 °C and 10 s⁻¹ yields a viscosity specification of 1.0–3.0 Pa·s; the upper limit is fixed by manual injection force through a 21G × 1.5 in needle, which must remain below 50 N to avoid depot injection failure. Upon contact with tissue fluid, the high-molecular-weight 85:15 PLGA solution undergoes phase inversion, with solvent exchange producing a porous depot shell within 60 s and a continuous drug reservoir matrix. The terminal product is a subcutaneous in situ forming depot intended for 1-month to 3-month release, depending on polymer concentration and depot geometry. In vitro release testing uses a USP Apparatus 4 flow-through system with pH 7.4 phosphate buffer at 37 °C and sink conditions maintained below 10% drug saturation. Residual N-methyl-2-pyrrolidone in the finished polymer matrix is controlled below 530 ppm under ICH Q3C Class 2, Option 2, corresponding to a permitted daily exposure of 5.3 mg/day. Compliance testing includes USP <71>, USP <85>, USP <788> for particulates after in vitro gelation, ISO 10993-6 for local tissue reaction, and ISO 10993-11 for systemic toxicity. A critical process limitation is the sensitivity of the ternary phase boundary to water uptake; if the polymer solution absorbs moisture above 0.15% w/w during syringe filling, the solution can undergo premature phase inversion and plug the 0.45 µm filter. Filling is therefore performed under ≤5% relative humidity with a post-fill polymer moisture test by Karl Fischer. The ratio of NMP to benzyl alcohol is also constrained: benzyl alcohol above 8% w/w accelerates the initial burst release beyond 20% of loaded drug in the first 24 h, while below 3% w/w the formulation becomes difficult to inject through the 21G needle at refrigerator temperature.

For intravitreal delivery, a 7.5% w/w solution of RESOMER RG 858 S in ethyl acetate:ethanol 90:10 is charged with 20% w/w dexamethasone based on total solids and cast onto a fluoropolymer-lined glass plate at 0.5 mm wet thickness. Solvent evaporation is conducted in a nitrogen-purged oven at 40 °C for 24 h, with a second-stage vacuum dry at 0.1 mbar for 12 h to remove residual ethyl acetate below 5000 ppm. The dried film is micro-cut into 0.5 mm × 0.5 mm × 6 mm rods and inspected for edge defects by optical microscopy at 20×. The terminal product is a biodegradable intravitreal insert that releases dexamethasone through diffusion and polymer bulk erosion, with the 85:15 lactide:glycolide ratio selected to extend the erosion phase beyond 3 months. In vitro release testing uses pH 7.4 phosphate-buffered saline with 0.1% w/v cetyltrimethylammonium bromide at 37 °C to maintain sink conditions for the lipophilic steroid. Compliance includes ISO 10993-5 for retinal pigment epithelial cytotoxicity, ISO 10993-10 for ocular irritation, USP <71>, USP <85>, and ICH Q3C residual solvent limits for ethyl acetate and ethanol. Gamma irradiation is not used for this insert because the radiation-induced reduction in polymer molecular weight alters the 3-month erosion profile; aseptic manufacturing with 0.22 µm filter sterilization of the casting solution and ethylene oxide terminal sterilization of the final insert per ISO 11135 are applied for sterility assurance. A key operational boundary is film brittleness at low humidity: below 30% relative humidity the dried PLGA film can fracture during micro-cutting, so the cutting station is maintained at 45–55% relative humidity and 20 °C. The final insert is packaged under nitrogen in foil pouches to restrict moisture uptake below 0.2% w/w over shelf life.

Does Gamma Irradiation Shift the Molecular Weight Distribution of RG 858 S Coatings?

Ultrasonic spray coating of a peripheral or coronary device is performed with a 1.5% w/w RESOMER RG 858 S solution in acetone:chloroform 70:30, containing sirolimus at 0.6% w/w based on total solution mass. A Sono-Tek ultrasonic nozzle operating at 120 kHz deposits the coating at 0.15 mL/min onto a bare cobalt-chromium L605 stent mounted on a rotating mandrel at 120 rpm. The process is run in an ISO 14644-1 Class 7 cleanroom with solvent vapor extraction. Coating thickness is controlled at 4.0 µm ± 0.5 µm by laser confocal microscopy, and coating uniformity is verified by scanning electron microscopy after balloon expansion to 3.0 mm nominal diameter. The terminal product is a drug-eluting stent with a bioresorbable PLGA carrier layer. In vitro drug release uses USP Apparatus 7 with reciprocating holders in pH 7.4 phosphate buffer containing 0.4% w/v sodium dodecyl sulfate at 37 °C. Biocompatibility for cardiovascular contact follows ISO 25539-2 for endovascular prosthesis requirements, ISO 10993-4 for hemocompatibility, ISO 10993-6 for vascular implantation, and ISO 10993-5 for cytotoxicity. Residual chloroform is controlled below 60 ppm under ICH Q3C Class 2 Option 2, and residual acetone is controlled below 5000 ppm. The molecular weight of the 85:15 PLGA carrier is measured by size-exclusion chromatography before and after terminal gamma irradiation. Published data for this specific configuration is limited; 25 kGy gamma sterilization can reduce the number-average molecular weight by chain scission, increasing the elution rate, so a dose range of 15–25 kGy with absorbed-dose mapping is specified. The coating is incompatible with amine-containing stabilizers because residual amine moieties accelerate ester hydrolysis during shelf storage. Storage is specified at 2–8 °C in moisture-barrier packaging with desiccant to keep package headspace humidity below 10% RH.

At bone void sites, a solvent-casting and particulate-leaching process uses 8% w/w RESOMER RG 858 S in chloroform, 80% w/w sodium chloride crystals sieved to 200–400 µm, and 3% w/w vancomycin hydrochloride relative to the dry scaffold mass. The paste is cast into a PTFE mold, lyophilized at 0.05 mbar for 24 h, and leached in deionized water for 48 h with five complete water changes. The terminal product is an antibiotic-eluting void filler with interconnected pores between 200 µm and 400 µm, intended for short-term local infection prophylaxis after debridement. In vitro antibacterial activity is verified by agar diffusion and broth dilution using Staphylococcus aureus ATCC 29213, with release sampling in pH 7.4 simulated body fluid at 37 °C. Compliance includes ISO 10993-1, ISO 10993-5, ISO 10993-6, USP <71>, USP <85>, and residual chloroform controlled below 60 ppm under ICH Q3C. The process limitation is the residual sodium chloride content; incomplete leaching above 0.5% w/w chloride raises the local osmotic pressure and delays osteoblast attachment in vitro. Published data for this specific formulation is limited; pore architecture and drug release are highly dependent on porogen particle size distribution and on the hydration state of the PLGA during casting.