Blending process modeling and control by multivariate curve resolution.

The application of the Multivariate Curve Resolution by Alternating Least Squares (MCR-ALS) method to model and control blend processes of pharmaceutical formulations is assessed. Within the MCR-ALS framework, different data analysis approaches have been tested depending on the objective of the study, i.e., knowing the effect of different factors in the evolution of the blending process (modeling) or detecting the blend end-point and monitoring the concentration of the different species during and at the end of the process (control). Data analysis has been carried out studying multiple blending runs simultaneously taking advantage of the multiset mode of the MCR-ALS method. During the ALS optimization, natural constraints, such as non-negativity (spectral and concentration directions) have been applied for blend modeling. When blending control is the main purpose, a variant of the MCR-ALS algorithm with correlation constraint in the concentration direction has been additionally used. This constraint incorporates an internal calibration procedure, which relates resolved concentration values (in arbitrary units) with the real reference concentration values in the calibration samples (known references) providing values in real concentration scale in the final MCR-ALS results. Two systems consisting of pharmaceutical mixtures of an active principle (acetaminophen) with two or four excipients have been investigated. In the first case, MCR results allowed the description of the evolution of the individual compounds and the assessment of some physical effects in the blending process. In the second case, MCR analysis allowed the detection of the end-point of the process and the assessment of the effects linked to variations in the concentration level of the compounds.

Near-infrared spectroscopy and imaging for the monitoring of powder blend homogeneity.

In-process testing requirements for adequacy of mixing are established in 21 CFR 211.110(a)(3). Considering also, the U.S. Food and Drug Administration's draft guidance published in 1999 (Guidance for Industry, ANDAs: Blend Uniformity Analysis; http://www.fda.gov/cder/guidance/index.htm), the importance of when and how to perform blend uniformity analysis is obvious. Near-infrared (NIR) spectroscopy was used noninvasively, in this study, to monitor powder blend homogeneity. Powder mixtures consisting of salicylic acid and Fast-Flo lactose were blended in an 8-qt. V-Blender. Optical ports installed at six positions on the blender allowed spectral collection using fiber optics. A traditional thief probe was used to collect powder samples for ultraviolet (UV) reference analysis. The blender was stopped at preselected time points for collection of NIR and UV data. Several algorithms and sampling protocols were studied to identify an optimum methodology for blend homogeneity determination. The blending process was also monitored with an InSb imaging camera for comparison with the traditional NIR spectroscopy and UV reference data. Data analysis indicates that multiple sampling points were essential for accurate and precise estimation of mixing end points. Moreover, multiple runs of identical blends often display homogeneity at unique end points, thus demonstrating the potential advantage of monitoring every blend.

Nondestructive tablet hardness testing by near-infrared spectroscopy: a new and robust spectral best-fit algorithm.

A new algorithm using common statistics was proposed for nondestructive near-infrared (near-IR) spectroscopic tablet hardness testing over a range of tablet potencies. The spectral features that allow near-IR tablet hardness testing were evaluated. Cimetidine tablets of 1-20% potency and 1-7 kp hardness were used for the development and testing of a new spectral best-fit algorithm for tablet hardness prediction. Actual tablet hardness values determined via a destructive diametral crushing test were used for construction of calibration models using principal component analysis/principal component regression (PCA/PCR) or the new algorithm. Both methods allowed the prediction of tablet hardness over the range of potencies studied. The spectral best-fit method compared favorably to the multivariate PCA/PCR method, but was easier to develop. The new approach offers advantages over wavelength-based regression models because the calculation of a spectral slope averages out the influence of individual spectral absorbance bands. The ability to generalize the hardness calibration over a range of potencies confirms the robust nature of the method.

Near-infrared spectroscopic characterization of pharmaceutical powder blends.

Near-infrared (near-IR) spectroscopy was used to qualitatively assess the homogeneity of a typical direct compression pharmaceutical powder blend consisting of hydrochlorothiazide, fast-flo lactose, croscarmellose sodium, and magnesium stearate. Near-IR diffuse reflectance spectra were collected from thieved powder samples using a grating-based spectrometer. A second-derivative calculation and principal component analysis were performed on the spectra prior to qualitative evaluation. Blend homogeneity was determined using single- and multiple-sample bootstrap algorithms and traditional chi-square analysis. The results suggested that bootstrap techniques provided greater sensitivity for assessing blend homogeneity than chi-square calculations and that near-IR has great potential as an analytical tool in powder blend analysis.

Near-infrared spectroscopic monitoring of the film coating process.

PURPOSE: The purpose of this study was to investigate the potential of near-infrared (near-IR) spectroscopy for non-destructive at-line determination of the amount of polymer coat applied to tablet cores in a Wurster column. METHODS: The effects of coating composition on the near-IR spectroscopic determination of ethylcellulose (Aquacoat ECD-30) or hydroxypropylmethylcellulose (HPMC)-based (Spectrablend) coating were evaluated, as were the performance of several chemometric techniques. RESULTS: Tablets were coated with up to 30% ethylcellulose or 22% HPMC, and samples were pulled at regular intervals during each coating run. Near-IR reflectance spectra of the intact tablets were then collected. The spectra were preprocessed by multiplicative scatter correction (MSC) or second derivative (D2) calculations, and calibrations developed using either principal components (PCs) or multiple spectral wavelengths. The near-IR method provided predictions of film applied with standard errors of 1.07% w/w or less. CONCLUSIONS: Near-IR spectroscopy can be profitably employed in a rapid and non-destructive determination of the amount of polymer film applied to tablets, and offers a simple means to monitor the film coating process.

Determination of film-coated tablet parameters by near-infrared spectroscopy.

Near-infrared (near-IR) spectroscopy was used in the determination of three parameters of theophylline tablets film-coated with ethylcellulose. Spectra of individual intact tablets were collected on two near-IR spectrometers: a grating-based spectrometer, and an acousto-optic tunable filter spectrometer. Calibrations were developed for the prediction of the time to 50% dissolution (t50%) of theophylline for tablets of varying coat thickness, for the determination of the thickness of the ethylcellulose coat applied, and for the prediction of the hardness of coated tablets. Principal component analysis was performed on the spectra prior to calibration development. The standard errors of calibration (SEC) and prediction (SEP) for determination of dissolution rates were 2.8 and 6.6 min, respectively. The SEC for the coating thickness calibration was 0.0002 inches, with an SEP of 0.00024 inches, and the SEC and SEP for the determination of tablet hardness were 0.54 and 0.62 kilopons, respectively.

Spectrophotometric prediction of the dissolution rate of carbamazepine tablets.

A near-infrared (IR) spectrophotometer, integrating optics, and parallel-vector supercomputer are employed to develop a mathematical model that predicts the dissolution rate of individual intact tablets from near-IR spectra (r2 = 0.985). Each tablet can be analyzed nondestructively by the spectrophotometer in less than 1 min. The model permits hundreds of near-IR wavelengths to be used in the determination of dissolution rate, leading to increased accuracy.

Near-infrared spectrometry of microorganisms in liquid pharmaceuticals.

Biotechnology and pharmaceutical research have created a number of new and potentially life-saving drugs. Many of these drugs are formulated as injectable products. Some drug products do not survive autoclaving or other means of terminal sterilization. An aseptic filling process is typically used to sterilize such products, but it is less reliable than autoclaving, making detection of unsterile units even more essential. Invasive microbiological methods and turbidimetry are currently employed as inspection techniques. These processes are time-consuming, destroy product, and may not detect low levels of contamination. Near-IR light scattering is proposed as a new method of determining low levels of contamination noninvasively and nondestructively. The method is used successfully in the current study to detect contamination by a species of yeast, mold, and bacteria in intact plastic infusion bags at levels as low as three colony-forming units per milliliter for yeast. By use of the near-IR method, each injectable unit can be evaluated with its integrity maintained, allowing the product to be dispensed or evaluated by another analytical method.

Nondestructive near-infrared analysis of intact tablets for determination of degradation products.

Near-infrared spectrometry was used in this study to examine intact aspirin tablets in order to demonstrate the usefulness of the technique as a nondestructive method of quality control. Unique sampling optics were used to simultaneously illuminate the entire surface of the tablets, including the top, bottom, and side. Changes in individual tablet spectra were correlated to (a) the time that the tablets spent in a hydrator, (b) the mass of water absorbed by the tablets, and (c) the mass of salicylic acid formed by base-catalyzed hydrolysis of acetylsalicylic acid. A prediction equation for each of these three parameters was constructed using near-infrared spectral reflectance values obtained from intact tablets. Prediction errors were low for (a) the time that tablets spent in the hydrator (+/- 19 h over a period of 168 h), (b) the mass of water absorbed (+/- 0.04% of tablet mass), and (c) the mass of salicylic acid formed (+/- 0.04% of tablet mass).