Application of reverse engineering in the field of pharmaceutical tablets using Raman mapping and chemometrics

Raman micro-spectroscopy technique offers a combination of relatively high spatial resolution with identification of components or mixtures of components in different sample areas, e.g. on the surface or the cross-section of a sample. This study is focused on the analysis of the tablets from pharmaceutical development with different technological parameters: (1) the manufacturing technology, (2) the particle size of the input API (active pharmaceutical ingredient) and (3) the quantitative composition of the individual excipients. These three mentioned parameters represent the most frequently solved problems in the field of reverse engineering in pharmacy. The investigation aims to distinguish tablets with the above-described technological parameters with limited subjective steps by Raman microscopy. Furthermore, non-subjective methods of Raman data analysis using advanced statistical analysis have been proposed, namely Principal Component Analysis, Soft Independent Modelling of Class Analogy and Linear Discriminant Analysis. The methods successfully distinguished and identified even very small differences in the analysed tablets within our study and provided objective statistic evaluation of Raman maps. The information on component and particle size distribution including their small differences, which is the critical parameter in the development of the original and generic products, was obtained due to combination of these methods. Even though each of these chemometric methods evaluates the data set from a different perspective, their mutual application on the problem of Raman maps evaluation confirmed and specified results on level that would be unattainable with the use of only one them. (c) 2021 Elsevier B.V. All rights reserved.

Automated summary

This study focuses on analyzing pharmaceutical tablets with different technological parameters using Raman micro-spectroscopy technique. Advanced statistical analysis methods such as Principal Component Analysis, Soft Independent Modelling of Class Analogy, and Linear Discriminant Analysis successfully distinguished and identified small differences in the tablets, providing objective statistic evaluation of Raman maps. The combination of these methods enabled the researchers to obtain critical information on component and particle size distribution.