Design of experiments methodology in studying near-infrared spectral information of model intact tablets : simultaneous determination of metoprolol tartrate and hydrochlorothiazide in solid dosage forms and powder compressibility assessment using near-infrared spectroscopy

Abstract

Near-infrared spectroscopy (NIRS) is applied in pharmaceutical industry for monitoring drug content during tablet manufacturing process. NIRS method, once developed and validated, is used over years and it is of critical importance to insure method robustness towards formulation, process, instrumental, acquisition and environmental factors. Design of Experiments (DoE) methodology was proposed in this work for systematic study of the effect of compression pressure, pre-compression pressure and tableting speed on Average Euclidean Distance (AED) which reflects NIR spectral features of the studied caffeine tablets, and Root Mean Squared Error of Prediction (RMSEP) as a key performance indicator of the developed NIRS calibration model for caffeine content prediction. Study was performed in diffuse reflectance (DR) and diffuse transmittance (DT) measurement mode. Tableting factors shown to have significant influence on the studied responses have been considered in the development of the robust calibration models in DR and DT mode, using Global Calibration Model (GCM) approach. Three studied factors have shown to be significant in DR mode whereas, compression pressure and tableting speed have shown significant effect on the studied responses in DT mode. Developed robust method in DT mode have shown superior performances compared to DR mode, exhibiting total error (RMSEP) of 1.21 % calculated on the independent test set. DoE setup, with the selection of factors and responses adopted in this study was not reported elsewhere. Simultaneous NIRS quantification of two APIs in powders and tablets requires several challenges to be overcome. Overlapping absorption peaks of formulation components result in method specificity problem. Strategy for selecting the samples used for developing the prediction models is needed. Robustness of the method towards formulation factors needs to be assessed due to complex formulation. Fast and simple method for simultaneous quantification of Hydrochlorothiazide (HTZ) and Metoprolol Tartrate (MTP) in powders and tablets was proposed in work. Simulation of industrial scale tablet machine using tablet press replicator - Presster® was proposed as fast and cost-effective alternative for design and manufacture of tablet sets needed for NIRS calibration model development. Balance Reference Method (BRM) was proposed as an alternative to HPLC and UV-spectroscopy which are traditionally used as reference methods in NIRS model development. The proposed experimental setup was suggested for the feasibility study stage of the method development. The two model drugs were simultaneously quantified using NIRS exhibiting RMSEP of 1.69 and 1.31 mg in HTZ powder and tablet samples respectively, while MTP powder and tablet samples were predicted with RMSEP of 3.15 and 3.00 mg respectively. NIRS analysis of Metoprolol Tartrate and Hydrochlorothiazide in powders and tablets was not yet reported in the literature. The compressibility and compatibility of a powder formulation is conventionally determined by compaction followed by destructive tensile strength and relative density measurement of the final compact. In this study, a non-destructive near-infrared spectroscopic (NIRS) was evaluated for the determination of powder compressibility and compactibility. Twelve different formulations were investigated with 2 batches produced per formulation. Relative density and tensile strength were measured using a traditional, destructive method on one tablet batch and subsequently by a developed non-destructive chemometric NIRS method on the second batch of the particular formulation. The outcomes of the two approaches were compared to validate the developed method. All data sets were fitted to the three established mathematical equations to calculate equation factors, which represent a formulation compressibility and compactibility. The study focus was set on the equation factor comparison between the traditional and the newly designed method. The results have shown a high degree similarity between the outcomes of the two methods. A discrepancy between the two methods was observed for the outcomes of the equation factors after fitting to Leuenberger equation. The approach using NIRS is suggested as a promising tool for monitoring tablet manufacturing process.