Purpose: Commonly available PLGA microspheres used for sustained release of hydrophobic drugs have compact matrix type morphology. Unlike this prevailing concept, this study aimed at applying porous PLGA microspheres with sponge-like skeleton to the development of sustained depot injectables of hydrophobic drugs.
Methods: Porous sponge-like PLGA microspheres laden with a model hydrophobic drug were prepared through a simple emulsion technique using a non-halogenated isopropyl formate and ammonia. Briefly, isopropyl formate, was used as a dispersed solvent to dissolve PLGA with a lactide:glycolide ratio of 75:25. A dispersed phase consisting of 4 ml of the green solvent, PLGA (0.25 - 0.45 g), and progesterone (40 - 80 mg) was emulsified in 40 ml of a 0.5% polyvinyl alcohol aqueous solution. When a 28% ammonia solution (3 to 7 ml) was added into the oil-in-water emulsion, ammonia partitioned into the emulsion droplets and reacted with isopropyl formate to produce water-soluble isopropanol and formamide. This ammonolysis reaction was complete after 7 min. Because they were anti-solvents toward PLGA, oil emulsion droplets were quickly transformed into solid microspheres. Also, their leaching into the aqueous phase left numerous pores across microsphere matrices. The microspheres were separated by filtration and redispersed in 150 ml of a 0.1% polyvinyl alcohol aqueous solution for 2 hrs. The resultant spongelike microspheres were collected by filtration and vacuum dried. As a control, non-porous compact microspheres were prepared by a typical solvent evaporation process. The major quality attributes of these microspheres evaluated in this study included microsphere morphology, drug encapsulation efficiency, thermal behavior, particle size distribution, and in vitro drug release patterns.
Results: Our microencapsulation technique enabled us to produce very impressive sponge-like PLGA microsphere where numerous pores and voids were ubiquitous (Figure 1). When non-porous microspheres were prepared according to the conventional solvent evaporation process, a significant portion of progesterone crystallized in the aqueous phase during the microencapsulation process. As shown in Figure 2A, particle sizes corresponding to progesterone crystals were observed when the microsphere suspension was subject to the particle size analysis. As a result, poor drug encapsulation efficiencies were observed. For example, when 0.25 g of PLGA and 40 mg of progesterone was used, the drug encapsulation efficiency was 54.0 +/- 6.2%. By sharp contrast, when the ammonolysis-based microencapsulation process was used to prepare spongy microspheres, the aforementioned drug crystallization phenomenon was completely suppressed (Figure 2B). When the same batch formula as above was used, the drug encapsulation efficiency was significantly increased to 83.1 +/- 6.0%. In our process, the microsphere hardening process occurred very quickly, which were able to overcome the drug nucleation and crystallization in the aqueous phase. Also, it was easy to adjust the residual content of isopropyl formate below 0.5% by simple vacuum drying. The extraordinary porosity of our sponge-like microspheres might have helped minimize the residual solvent level to less than 0.5%. Finally, the in vitro release study demonstrated that our microspheres were capable of releasing progesterone continuously for 2 weeks without lag time.
Conclusion: Most PLGA microspheres used as sustained release depots for hydrophobic drugs have dense, compact matrix structure. While making these PLGA microspheres, drug crystallization often occurs. In addition, due to their compact matrix structure, it is not easy to completely remove a dispersed organic solvent from the hardened microspheres. Finally, a lag time is often required prior to full-fledged drug release. Our spongelike microspheres are anticipated to be an excellent alternative to overcome these limitations.
Seoyeon Kim– Graduate student, Ewha Womans University, Seoul, Seoul-t'ukpyolsi