Topical liposomes treated by probe-sonication. Study on process and composition using statistical experimental design and multivariate evaluation

Abstract

Abstract A size distribution between 200 and 300 nm and a high drug entrapment is desirable for liposomes intended for topical application. In an early phase of drug development, small batch sizes are also wanted. Dual Asymmetric Centrifugation (DAC) meets these requirement and was our method of choice. But, after successful production of a few DAC-liposome batches with drug entrapment of chloramphenicol (CAM) > 50%, vesicle size 200-300 nm, batch size approx. 500 µl, high lipid content of 40-50% (w/v), the DAC-machine got broken. Thus, probe-sonication was replacing DAC, with the following alterations: (1) The sample volume was increase from 500 µl to between 2 and 8 ml (2) Lipid concentration was reduced from semisolid VPGs to liquid dispersions; 10, 20 and 30 mg/ml lipid conc. in sonicated samples. The efficiency of the DAC method was significantly improved when adding propylene glycol (PG), reducing the needed processing time from 50 min to 2 min and 55 seconds. We therefore wanted to investigate (3) the effect of PG-concentration on the sonication method (levels of between 50 and 200 µl PG was added). Finally, (4) number of 2-minutes sonication cycles was judged critical, and between 2 and 6 cycles were tested. Empty liposomes were produced from Lipoid S-100 and variables investigated using multifactorial design on two levels (+1/-1), plus center points. The design matrix, given by the Unscramble 9.8 software, gave 19 experiments in a (24) full factorial design. The liposome size distribution varied from 55.8 to 408 nm, except one experiment giving liposome sizes of 876 nm. The ANOVA analysis suggest that within the levels of the variables investigated, sample volume is the most important variable affecting the vesicle size (p-value 0.0145), and number of sonication cycles (p-value 0.0692). Of the 19 experiments, four experiments had the aimed vesicle size of 200-300 nm, and were repeated with CAM added. CAM entrapped in sonicated liposomes was between 23 and 31%, and lower that for the DAC-liposomes (around 50%). An increase in vesicle size was observed when adding CAM for three out of four formulations (mean diameter 769, 837 and 834 nm and 67 nm). Both DAC and probe-sonication are suitable for making liposomes with the aimed vesicle size of 200-300 nm, and with acceptable incorporation efficiency of CAM. For CAM-liposomes, sonication conditions applied in this experimental matrix is too gentle and liposome size bigger than aimed for. Effect of PG on liposome size from sonication could not be demonstrated with the applied PG concentrations.

Keywords: Drug delivery system; liposome; probe sonication; multivariate analysis; statistical experimental design, Dual asymmetric centrifugation, Photon correlation spectroscopy