Optimization of a high strenght mesalamine multi-particulate dosage form

Background: 5-aminosalicylic acid (5-ASA, mesalamine) is at present the treatment of choice for Inflammatory Bowel Diseases (IBD), with a therapeutic regimen of 2.4-4.8 g/day. However, most of the mesalamine-based products currently available on the market contain a dose of 5-ASA (generally 500 mg API/unit) which is relatively low if considering the aforementioned dosage regimen. A high strength dosage form containing more than 1 g 5-ASA would therefore be of interest to reduce daily administrations and hence improving compliance. Multi-particulates dosage forms, based on high density pellets, have been sought as promising alternatives for the administration of 5-ASA, both for their biopharmaceutical (e.g. more even and predictable distribution and transportation in the gastro-intestinal tract) and technological (e.g. high drug loading and ease of drug release modulation) advantages. In a previous study (Di Pretoro et al., 2010) a lab-scale ram extruder was employed to investigate the influence of 5-ASA chemical (acidity) and physical (particle size and shape) characteristics on the rheological behaviour of highly-drug loaded pastes. A formulation route for the development of a 90 wt% 5-ASA/microcrystalline cellulose wet mass based on milled 5-ASA, which gave good yields of pellets at the lab scale, was successfully identified. Purpose: To improve further the promising formulation identified at the lab scale and to assess the feasibility of scaling up to a pilot plant apparatus (basket extruder) using a mixed fractional factorial approach. In addition, an optimisation study, aiming to identify the optimal formulation and process parameters for the development of a high strength multi-particulate dosage form at the pilot scale, was performed. Methods: A Nica\uae pilot-scale extruder (basket type), coupled with a cross-hatched friction plate spheroniser, was used for pellet preparation. Colloidal microcrystalline cellulose (Avicel RC591) was chosen as E-S aid. The primary liquid binder was water, with polyvinylpyrrolidone (PVP) used as extra binder. A 3x24-1 fractional factorial design was employed to investigate the effects of three formulation components (5-ASA loading, PVP content and water amount in paste formulation) and two process parameters (extrusion speed and spheronisation time) on process yield and on the final properties of the pellets, namely size, shape, mechanical resistance, bulk density and dissolution properties. 27 runs, including 3 replicates of the central point, were performed. An optimisation process was also carried out, using the desirability function. Results: The 5-ASA loading, PVP content and water amount had a significant effect (p