A Predictive Computational Analytical Chemistry Approach for Quantifying Drug–Excipient Interactions Using FTIR Spectral Moments

Abstract

Active pharmaceutical ingredient and excipient interactions play a direct decisive role in the stability, safety and quality of pharmaceutical formulations and they are generally examined based on qualitative observations in the literature. This study presents an original computational analytical chemistry framework that combines ATR-FTIR spectral moments with stochastic modeling, implemented for the first time in pharmaceutical literature. In the scope of the research, the interaction potential between two widely used model excipients, β-cyclodextrin and glycine, were calculated with an approach investigating spectral changes. Spectral data were parametrically defined through mathematical moments such as area, centroid (μ), bandwidth (σ), and skewness and these parameters were converted into an "Interaction Degree" (ID) metric. To test analytical reliability and model robustness, uncertainty propagation and sensitivity analyses were performed using Monte Carlo simulations involving 50,000 iterations.A closed form predictive model developed over the obtained data was found to have a very high coefficient of determination (R2) of 0.978. This highly value proved that there is an excellent agreement between simulated data and theoretical predictions. As a result of sensitivity analyses, it was determined that band asymmetry, which is generally ignored in the literature, is the most dominant and critical parameter in determining the interaction degree. In conclusion, this study provides a predictive methodology that goes beyond traditional methods for the fast, precise, and quantitative determination of excipient compatibility in pharmaceutical quality control processes.