Mechanistic evaluation of the initial burst release of leuprolide from spray-dried PLGA microspheres

Spray-dried poly(lactic-co-glycolic acid) (PLGA) peptide-loaded microspheres have demonstrated similar long-term in vitro release kinetics compared to those produced by the solvent evaporation method and commercial products. However, the difficult-to-control initial burst release over the first 24 h after administration presents an obstacle to product development and establishing bioequivalence. Currently, detailed information about underlying mechanisms of the initial burst release from microspheres is limited. We investigated the mechanism and extent of initial burst release using 16 previously developed spray-dried microsphere formulations of the hormone drug, leuprolide acetate, with similar composition to the commercial 1-month Lupron Depot® (LD). The burst release kinetics was measured with a previously validated continuous monitoring system as well as traditional sample-and-separate methods. The changes in pore structure and polymer permeability were investigated by SEM imaging and the uptake of a bodipy-dextran probe. In vitro results were compared to pharmacokinetics in rats over the same interval. High-burst, spray-dried microspheres were differentiated in the well-mixed continuous monitoring system but reached an upper limit when measured by the sample-and-separate method. Pore-like occlusions observed by confocal microscopy in some formulations indicated that particle swelling may have contributed to probe diffusion through the polymer phase and showed the extensive internal pore structure of spray-dried particles. Continuous monitoring revealed a rapid primary (1°) phase followed by a constant-rate secondary (2°) release phase, which comprised ∼80% and 20% of the 24-hr release, respectively. The ratio of 1° phase duration (t1°) and the characteristic probe diffusion time (τ) was highly correlated to 1° phase release for spray dried particles. Of the four spray-dried formulations administered in vivo, three spray-dried microspheres with similar polymer density showed nearly ideal linear correlation between in vivo absorption and well-mixed in vitro release kinetics over the first 24 h. By contrast, the more structurally dense LD and a more-dense in-house formulation showed a slight lag phase in vivo relative to in vitro. Furthermore, in vitro dimensionless times (tburst/τ) were highly correlated with pharmacokinetic parameters for spray-dried microspheres but not for LD. While the correlation of increases in effective probe diffusion and 1° phase release strongly suggests diffusion through the polymer matrix as a major release mechanism both in vitro and in vivo, a fixed lower limit for this release fraction implies an alternative release mechanism. Overall, continuous monitoring release and probe diffusion appears to have potential in differentiating between leuprolide formulations and establishing relationships between in vitro release and in vivo absorption during the initial burst period.