Poly(lactide-co-glycolide) microsphere formulations of darbepoetin alfa: spray drying is an alternative to encapsulation by spray-freeze drying.

PURPOSE: The purpose of this work was to evaluate spray-freeze drying and spray drying processes for encapsulation of darbepoetin alfa (NESP, Aranesp). METHODS: Darbepoetin alfa was encapsulated in poly(lactide-co-glycolide) by spray-freeze drying and by spray drying. Integrity was evaluated by size-exclusion chromatography and Western blot. Physical properties and in vitro release kinetics were characterized. Pharmacokinetics and pharmacodynamics were evaluated in nude rats. RESULTS: Microspheres produced by spray drying were larger than those produced by spray-freeze drying (69 microm vs. 29 microm). Postencapsulation integrity was excellent for both processes, with < 2% dimer by size-exclusion chromatography. In vitro release profiles were similar, with low burst (< 25%) and low cumulative protein recovery at 4 weeks (< or = 30%), after which time covalent dimer (< or = 6.5%) and high molecular weight aggregates (< or = 2.3%) were recovered by denaturing extraction. After a single injection, darbepoetin alfa was detected in serum through 4 weeks for all microsphere formulations tested in vivo, although relative bioavailability was higher for spray-freeze drying (28%) compared with spray drying (21%; p = 0.08) as were yields (73-82% vs. 34-57%, respectively). For both processes hemoglobin was elevated for 7 weeks, over twice as long as unencapsulated drug. CONCLUSIONS: Spray drying, conducted at pilot scale with commercial equipment, is comparable to spray-freeze drying for encapsulation of darbepoetin alfa.

Protein powders for encapsulation: a comparison of spray-freeze drying and spray drying of darbepoetin alfa.

PURPOSE: To evaluate spray-freeze drying and spray drying processes for fabricating micron-sized particles of darbepoetin alfa (NESP, Aranesp) with uniform size distribution and retention of protein integrity, requirements for encapsulation. METHODS: Darbepoetin alfa was spray-freeze dried using ultrasonic atomization at 120 kHz and 25 kHz and spray dried at bench-top and pilot scales. Reconstituted powders were evaluated by size exclusion chromatography and UV/VIS spectroscopy. Powder physical properties were also characterized. RESULTS: Spray-freeze drying resulted in aggregation of darbepoetin alfa. Aggregates (primarily insoluble) formed on drying and/or reconstitution. Particle size distributions were broad (span > or = 3.6) at both nozzle frequencies. Annealing before drying reduced aggregate levels slightly but increased particle size over 5-fold. Spray drying at inlet temperatures up to 135 degrees C (and outlet temperatures up to 95 degrees C) showed little impact on integrity. Integrity at bench-top and pilot scales was identical, with 0.2% dimer and no high molecular weight or insoluble aggregates detected. Particle size was small (< or = 2.3 microm) with uniform distribution (span < or = 1.4) at both process scales. CONCLUSIONS: Under the conditions tested spray drying, conducted at bench-top and pilot scales with commercially available equipment, was superior to spray-freeze drying for the fabrication of darbepoetin alfa particles for encapsulation.

Carbon dioxide extraction of residual solvents in poly(lactide-co-glycolide) microparticles.

A process for the reduction of residual solvents in spray-dried poly(lactide-co-glycolide) (PLGA)-darbepoetin alfa microparticles was developed using carbon dioxide (CO(2)) as an extraction solvent. CO(2) was investigated in two phase states, liquid and gas. Detrimental effects on encapsulated protein integrity and microparticle morphology were observed with liquid CO(2) exposure. Extraction with CO(2) gas at <100 psig reduced residual solvent concentration and particle agglomeration was limited. Extraction rates and particle agglomeration increased with higher CO(2) gas pressures. The CO(2) pressures below which particles of polylactide (PLA) and PLGA microparticles significantly agglomerated were determined and the data used to develop extraction cycles. Extraction cycles were developed in which CO(2) gas pressure was increased as residual solvent concentration decreased in order to keep extraction rates high throughout the cycle. Spray dried darbepoetin alfa-PLGA microparticles were extracted with CO(2) gas and characterized for residual solvent concentration, process yield, particle size distribution, morphology, and protein integrity. The results indicated CO(2) gas may be used to reduce residual solvent to approximately 200 ppm with no significant detrimental effects on protein integrity or microparticle morphology.