Teneligtiptin-Loaded Nanosponges - A Novel Strategic Approach Utilizing Artificial Intelligence

Teneligliptin is a DPP-4 inhibitor used in the management of type 2 diabetes mellitus; it suffers from poor aqueous solubility and limited bioavailability.To overcome these challenges, nanosponges have emerged as a promising drug delivery system due to their high entrapment efficiency, controlled release capabilities, and enhanced solubility.This study aimed to develop Teneligliptin-loaded nanosponges and apply artificial intelligence, specifically the Random Forest Regression (RFR) algorithm, to model and predict key formulation outcomes.Nanosponges were prepared using the emulsion-solvent diffusion method with ethyl cellulose and β-cyclodextrin.Six formulations (F1-F6) were developed by varying polymer and complexing agent concentrations.The formulations were characterized for, entrapment efficiency (EE%), and drug release.The RFR model was trained using formulation variables as input and EE% and % as the target.Fourier Transform infrared (FT-IR) spectra studies represented no interaction between the drug and excipients.Among six formulations, formulation 6 (F6) was found to be an optimized formulation.The percentages of entrapment efficiency were determined to be 92%.In vitro drug release was about 95 % at the end of the 24 th hour.The optimized formulations were evaluated for particle size, zetapotential and SEM analysis.The optimized formulation follows zero-order kinetics.The results of the stability study revealed that the optimized formulations were stable during the stability study.Integration of AI through Random Forest Regression provided a reliable, time-saving, and predictive approach to optimize nanosponge-based drug delivery systems.This strategic combination of nanotechnology and machine learning offers a robust platform for enhancing the bioavailability of poorly soluble drugs like Teneligliptin.