« Back to Farmacia Journal 2/2025

DEVELOPMENT OF HIGH-THROUGHPUT NIR-CHEMOMETRIC METHODS FOR THE CHARACTERISATION OF POWDER BLENDS FOR TABLETTING AND EXTENDED-RELEASE TABLETS WITH KETOPROFEN

TIBOR CASIAN 1, ANDRA REZNEK 1, LUCIA MARIA RUS 2*, ANDREEA LOREDANA VONICA-ȚINCU 3, RAREȘ IOVANOV 1, IOAN TOMUȚĂ 1

1 Department of Pharmaceutical Technology and Biopharmacy, Faculty of Pharmacy, “Iuliu Hațieganu” University of Medicine and Pharmacy, 41 Victor Babeș Street, Cluj-Napoca, Romania
2 Department of Drug Analysis, Faculty of Pharmacy, “Iuliu Hațieganu” University of Medicine and Pharmacy, 6 Louis Pasteur Street, Cluj-Napoca, Romania
3 Department of Preclinical Medicine, Faculty of Medicine, “Lucian Blaga” University, 2A Lucian Blaga Street, Sibiu, Romania

Download Full Article PDF

This study aimed to develop and validate NIR-chemometric methods for direct quantification of ketoprofen and lactose in powder blends for tabletting, respectively, ketoprofen, lactose, and the kinetics of release from extended-release inert matrix tablets. Calibration samples were prepared using an experimental design with two variables and five levels, including 26 formulations. The in vitro release was carried out in phosphate buffer (pH = 7.4), utilising a dissolution testing device equipped with baskets at 100 rpm. The calibration algorithm for ketoprofen was obtained using as pre-treatment method FD+MSC (PLS = 7, RMSECV = 1.46) for powder blends, respectively SNV (PLS = 10, RMSECV = 3.94) for tablets in the 10000 - 6098 cm-1; 5454 - 4596 cm-1 spectral ranges. In the case of lactose, the calibration algorithm was obtained using the pre-treatment method SLS (PLS = 7, RMSECV = 1.79) for powder blends, respectively FD+SNV (PLS = 4, RMSECV = 6.45) for tablets, in the 10000 - 4200 cm-1 spectral range. For kinetic release characterization, the Higuchi equation parameters were determined and correlated with the tablets’ NIR spectra to develop a chemometric method to estimate the kinetic release directly. According to the experimental data, the calibration algorithms were validated with satisfactory accuracy, reproducibility and linearity.