Effect of PLGA raw materials on in vitro and in vivo performance of drug-loaded microspheres.

Poly(lactide-co-glycolide) and poly(lactic-co-glycolic acids) (PLGAs) play a critical role in the development of commercial long-acting injectable microsphere formulations. However, very little information is available describing the impact of PLGA manufacturer and monomer distribution along the polymer chain (e.g., glycolic blockiness (R) and average lactic block length (L)) on the degradation and release behavior of PLGA drug carriers in vitro and in vivo. Here, we compared the in vitro and in vivo performance of (a) four leuprolide-loaded microsphere formulations prepared from similar low-molecular-weight acid-capped PLGAs (10-14 kD, i.e., Expansorb DLG 75-2A, Purasorb PDLG 7502A, Resomer RG 752H and Wako 7515) and (b) two triamcinolone acetonide-loaded (Tr-A) microsphere formulations from similar medium-molecular-weight ester-capped PLGAs (i.e., Expansorb DLG 75-4E and Resomer RG 753S). Lupron Depot and Zilretta were used as reference commercial products. The six 75/25 PLGAs displayed block lengths that were either above or below values expected from a random copolymer. Drug release and polymer degradation were monitored simultaneously in vitro and in vivo using a cage implant system. The four leuprolide-loaded formulations showed similar release and degradation patterns with some notable differences between each other. Microspheres from the Expansorb polymer displayed lower L and higher R relative to the other 3 PLGA 75/25 microspheres, and likewise exhibited distinct peptide release and degradation behavior compared to the other 3 formulations. For each formulation, leuprolide release was erosion-controlled up to about 30% release after the initial burst followed by a faster than erosion release phase. In vitro release was similar as that in vivo over the first phase but notably different from the latter release phase, particularly for the most blocky Expansorb formulation. The Purasorb and Wako formulations displayed highly similar performance in release, degradation, and erosion analysis. By contrast, the two ester-capped Expansorb DLG 75-4E and Resomer RG 753S used to prepare Tr-A microspheres shared essentially identical L and higher R and behaved similarly although the Expansorb degraded and released the steroid faster in vivo, suggestive of other factors responsible (e.g., residual monomer). The in vivo release performance for both drugs from the six microsphere formulations was similar to that of the commercial reference products. In summary, this work details information on comparing the similarities and differences in in vitro and in vivo performance of drug-loaded microspheres as a function of manufacturing and microstructural variables of different types of PLGA raw materials utilized and could, therefore, be meaningful in guiding the source control during development and manufacturing of PLGA microsphere-based drug products. Future work will expand the analysis to include a broader range of L and higher R, and add additional important formulation metrics (e.g., thermal analysis, and residual monomer, moisture, and organic solvent levels).