Development of a Near Infrared Spectroscopy method for the in-line quantitative bilastine drug determination during pharmaceutical powders blending.
Bilastine
Chemometric
Near Infrared Spectroscopy
Process Analytical Technology (PAT)
Quantitative analysis
Spectra pre-treatment
Journal
Journal of pharmaceutical and biomedical analysis
ISSN: 1873-264X
Titre abrégé: J Pharm Biomed Anal
Pays: England
ID NLM: 8309336
Informations de publication
Date de publication:
10 Sep 2021
10 Sep 2021
Historique:
received:
26
02
2021
revised:
13
07
2021
accepted:
21
07
2021
pubmed:
1
8
2021
medline:
25
8
2021
entrez:
31
7
2021
Statut:
ppublish
Résumé
The Food and Drug Administration (FDA)'s guidelines and the Process Analytical Technology (PAT) approach conceptualize the idea of real time monitoring of a process, with the primary objective of improvement of quality and also of time and resources saving. New instruments are needed to perform an efficient PAT process control and Near Infrared Spectroscopy (NIRS), thanks to its rapid and drastic development of last years, could be a very good choice, in virtue of its high versatility, speed of analysis, non-destructiveness and absence of sample chemical treatment. This work was aimed to develop a NIR analytical method for bilastine assay in powder mixtures for direct compression. In particular, the use of NIR instrumentation should allow to control the bilastine concentration and the whole blending process, assuring the achievement of a homogeneous blend. The commercial tablet formulation of bilastine was particularly suitable for this purpose, due to its simple composition (four excipients) and direct compression manufacturing process. Calibration and validation set were prepared according to a Placket-Burman experimental design and acquired with a miniaturized NIR in-line instrument (MicroNIR by Viavi Solution Inc.). Chemometric was applied to optimize information extraction from spectra, by subjecting them to a Standard Normal Variate (SNV) and a Savitzky-Golay second derivative pre-treatment. This spectra pre-treatment, combined with the most suitable wavelength interval (resulted between 1087 and 1217 nm), enabled to obtain a Partial Least Square (PLS) model with a good predictive ability. The selected model, tried on laboratory and production batches, provided in both cases good assay predictions. Results were confirmed by traditional HPLC (High Performance Liquid Chromatography) API (Active Pharmaceutical Ingredient) content uniformity test on the final product.
Identifiants
pubmed: 34332309
pii: S0731-7085(21)00388-5
doi: 10.1016/j.jpba.2021.114277
pii:
doi:
Substances chimiques
Benzimidazoles
0
Piperidines
0
Powders
0
Tablets
0
bilastine
PA1123N395
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
114277Informations de copyright
Copyright © 2021 Elsevier B.V. All rights reserved.
Déclaration de conflit d'intérêts
Declaration of Competing Interest The authors report no declarations of interest.