Artificial Neural Network (ANN) Approach to Predict an Optimized pH-Dependent Mesalamine Matrix Tablet

Area of Science:

• Pharmaceutical Sciences
• Drug Delivery Systems
• Materials Science

Background:

• Ulcerative colitis complications like severe bleeding and perforation necessitate targeted drug delivery.
• Current colon-targeted delivery systems often involve complex formulations and pH-sensitive coatings.

Purpose of the Study:

• To develop dicalcium phosphate (DCP)-facilitated, Eudragit-S100-based, pH-dependent, uncoated mesalamine matrix tablets.
• To create a simpler, effective alternative to conventional coated colon-targeted drug delivery systems.

Main Methods:

• Mesalamine formulations were prepared using wet granulation with varying DCP and Eudragit-S100 ratios.
• Physicochemical properties and in-vitro drug release profiles were evaluated.
• Artificial Neural Network (ANN) was employed for formulation optimization.

Main Results:

• Tablets met all critical quality attributes, including thickness, hardness, weight variation, friability, and content uniformity.
• The optimized formulation exhibited controlled release, releasing 12.09% at 2h and 72.96% at 12h, fitting a complex release mechanism (Weibull model, b=1.3).
• No significant drug-excipient interactions were observed via infrared spectroscopy.

Conclusions:

• The DCP-Eudragit-S100 blend effectively controlled mesalamine release without the need for coating.
• This uncoated formulation shows promise for improved ulcerative colitis management.