Effect of lactide/glycolide ratio on the in vitro release of ganciclovir and its lipophilic prodrug (GCV-monobutyrate) from PLGA microspheres

PurposeThe aim of this study is to investigate the roles of the lactide/glycolide ratio and prodrug derivatization on the in vitro release of ganciclovir and its lipophilic prodrug (GCV-monobutyrate) from PLGA microspheres for the treatment of cytomegalovirus (CMV) retinitis.MethodTwo grades of PLGA, a higher lactide content PLGA 6535 [d,l-lactide:glycolide = 65:35, MW 45,000–75,000 Da] and lower lactide content PLGA 5050 [d,l-lactide:glycolide = 50:50, MW 45,000–75,000 Da] were employed to prepare GCV loaded microspheres. The effect of lipophilic prodrug derivatization was investigated by converting GCV to GCV-monobutyrate (GCVMB). Microspheres were prepared by the oil-in-oil (O/O) solvent evaporation method and characterized in vitro, by studying their surface/internal morphology, entrapment efficiency, particle size, drug release, true density and glass transition temperature. In vitro release data were analyzed by a model equation to estimate various parameters of the drug release curves.ResultsThe O/O solvent evaporation method generated a high drug payload of up to 91%. Higher entrapment efficiencies were observed in the case of hydrophilic drug (GCV) relative to the lipophilic prodrug (GCVMB). Loosely bound or surface adsorbed drug/prodrug molecules may have resulted in the very short period (about 6 h) of the initial burst phase in all types of microspheres. GCV loaded microspheres utilized more time to release 50% (T50 value) of entrapped drug than GCVMB microspheres. T50 values estimated for GCVMB were shorter than those for GCV from microspheres with similar lactide/glycolide ratios. Lactide content in PLGA did not significantly alter GCVMB release relative to GCV release. The proposed model equation effectively estimated the drug release parameters (R2 > 0.98) with all drug/prodrug-PLGA combinations. SEM pictures have revealed that although both GCV and GCVMB microspheres were spherical but internal morphology was different, with former having uniform and dense whereas later have porous structures. Corroborating with internal morphologies, results revealed that true densities of GCV microspheres were relatively greater than corresponding GCVMB microspheres.ConclusionThe proposed method of preparation yields higher efficiency of drug entrapment for the hydrophilic drug. Prodrug entrapment into microspheres could result in longer residence time at the site of administration due to multiple processes involved in drug release at infected tissue. These processes include release from microspheres and enzymatic conversion of the prodrug to parent drug.