Monitoring of a film coating process for tablets using near infrared reflectance spectrometry

A process analytical chemical method using near infrared diffuse reflectance spectrometry was developed for the determination of the amount of tablet coating on single tablets. This method is based on calibration of the spectra versus the added mass of coating solution. The tablet core was composed of two halves of different chemical composition and spectra were recorded from both sides of the tablets. The calibration was carried out using the chemometric methods principal component analysis (PCA), partial least squares (PLS), and multiplicative signal correction (MSC). The PLS-model utilised spectra obtained from both sides, pretreated with MSC, and ordered into one object. This method can be used in process analytical chemistry at-line. Additional characterisation of the measurements was obtained by calibrating the spectra versus coating thicknesses obtained from optical microscopy. Using PCA, it was possible to roughly estimate the maximum depth in the coating material that returns chemical information, the 'information depth', which was 0.1-0.2 mm.     

Assessment of the beryllium lymphocyte proliferation test using statistical process control

Despite more than 20 years of surveillance and epidemiologic studies using the beryllium blood lymphocyte proliferation test (BeBLPT) as a measure of beryllium sensitization (BeS) and as an aid for diagnosing subclinical chronic beryllium disease (CBD), improvements in specific understanding of the inhalation toxicology of CBD have been limited. Although epidemiologic data suggest that BeS and CBD risks vary by process/work activity, it has proven difficult to reach specific conclusions regarding the dose-response relationship between workplace beryllium exposure and BeS or subclinical CBD. One possible reason for this uncertainty could be misclassification of BeS resulting from variation in BeBLPT testing performance. The reliability of the BeBLPT, a biological assay that measures beryllium sensitization, is unknown. To assess the performance of four laboratories that conducted this test, we used data from a medical surveillance program that offered testing for beryllium sensitization with the BeBLPT. The study population was workers exposed to beryllium at various facilities over a 10-year period (1992-2001). Workers with abnormal results were offered diagnostic workups for CBD. Our analyses used a standard statistical technique, statistical process control (SPC), to evaluate test reliability. The study design involved a repeated measures analysis of BeBLPT results generated from the company-wide, longitudinal testing. Analytical methods included use of (1) statistical process control charts that examined temporal patterns of variation for the stimulation index, a measure of cell reactivity to beryllium; (2) correlation analysis that compared prior perceptions of BeBLPT instability to the statistical measures of test variation; and (3) assessment of the variation in the proportion of missing test results and how time periods with more missing data influenced SPC findings. During the period of this study, all laboratories displayed variation in test results that were beyond what would be expected due to chance alone. Patterns of test results suggested that variations were systematic. We conclude that laboratories performing the BeBLPT or other similar biological assays of immunological response could benefit from a statistical approach such as SPC to improve quality management.     

Assessment of the Beryllium Lymphocyte Proliferation Test Using Statistical Process Control

Despite more than 20 years of surveillance and epidemiologic studies using the beryllium blood lymphocyte proliferation test (BeBLPT) as a measure of beryllium sensitization (BeS) and as an aid for diagnosing subclinical chronic beryllium disease (CBD), improvements in specific understanding of the inhalation toxicology of CBD have been limited. Although epidemiologic data suggest that BeS and CBD risks vary by process/work activity, it has proven difficult to reach specific conclusions regarding the dose-response relationship between workplace beryllium exposure and BeS or subclinical CBD. One possible reason for this uncertainty could be misclassification of BeS resulting from variation in BeBLPT testing performance. The reliability of the BeBLPT, a biological assay that measures beryllium sensitization, is unknown. To assess the performance of four laboratories that conducted this test, we used data from a medical surveillance program that offered testing for beryllium sensitization with the BeBLPT. The study population was workers exposed to beryllium at various facilities over a 10-year period (1992–2001). Workers with abnormal results were offered diagnostic workups for CBD. Our analyses used a standard statistical technique, statistical process control (SPC), to evaluate test reliability. The study design involved a repeated measures analysis of BeBLPT results generated from the company-wide, longitudinal testing. Analytical methods included use of (1) statistical process control charts that examined temporal patterns of variation for the stimulation index, a measure of cell reactivity to beryllium; (2) correlation analysis that compared prior perceptions of BeBLPT instability to the statistical measures of test variation; and (3) assessment of the variation in the proportion of missing test results and how time periods with more missing data influenced SPC findings. During the period of this study, all laboratories displayed variation in test results that were beyond what would be expected due to chance alone. Patterns of test results suggested that variations were systematic. We conclude that laboratories performing the BeBLPT or other similar biological assays of immunological response could benefit from a statistical approach such as SPC to improve quality management.     

IR spectroscopy together with multivariate data analysis as a process analytical tool for in-line monitoring of crystallization process and solid-state analysis of crystalline product

Crystalline product should exist in optimal polymorphic form. Robust and reliable method for polymorph characterization is of great importance. In this work, infra red (IR) spectroscopy is applied for monitoring of crystallization process in situ. The results show that attenuated total reflection Fourier transform infra red (ATR-FTIR) spectroscopy provides valuable information on process, which can be utilized for more controlled crystallization processes. Diffuse reflectance Fourier transform infra red (DRIFT-IR) is applied for polymorphic characterization of crystalline product using X-ray powder diffraction (XRPD) as a reference technique. In order to fully utilize DRIFT, the application of multivariate techniques are needed, e.g., multivariate statistical process control (MSPC), principal component analysis (PCA) and partial least squares (PLS). The results demonstrate that multivariate techniques provide the powerful tool for rapid evaluation of spectral data and also enable more reliable quantification of polymorphic composition of samples being mixtures of two or more polymorphs. This opens new perspectives for understanding crystallization processes and increases the level of safety within the manufacture of pharmaceutics.     

The synthesis of metoprolol monitored using Raman spectroscopy and chemometrics.

The synthesis of Metoprolol base was studied using Raman spectroscopy with a 785-nm laser, optical fibres, a holographic transmission grating, confocal optics and a charge-coupled device (CCD) detector. The reaction mixture was heated according to a temperature gradient and spectra of the reaction mixture were obtained by focusing the laser beam through ordinary reaction flasks. Because of overlapping bands, multivariate techniques such as principal components analysis (PCA) and partial least-squares projections to latent structures (PLS) were used in the evaluation of the obtained spectra. The use of PCA or PLS against time does not require any calibration samples and a quantitative calibration is not necessary in order to monitor the reaction. A method for reaction endpoint determination, based on euclidean distances in the score space, is presented. The use of multivariate batch control charts have been demonstrated and a number of problems and solutions regarding the sample presentation have been discussed. The effect of spectral pretreatment on the multivariate results is shown and discussed. The monitoring results show that the time to produce Metoprolol base could be reduced.     

Estimation of active pharmaceutical ingredients content using locally weighted partial least squares and statistical wavelength selection

Development of quality estimation models using near infrared spectroscopy (NIRS) and multivariate analysis has been accelerated as a process analytical technology (PAT) tool in the pharmaceutical industry. Although linear regression methods such as partial least squares (PLS) are widely used, they cannot always achieve high estimation accuracy because physical and chemical properties of a measuring object have a complex effect on NIR spectra. In this research, locally weighted PLS (LW-PLS) which utilizes a newly defined similarity between samples is proposed to estimate active pharmaceutical ingredient (API) content in granules for tableting. In addition, a statistical wavelength selection method which quantifies the effect of API content and other factors on NIR spectra is proposed. LW-PLS and the proposed wavelength selection method were applied to real process data provided by Daiichi Sankyo Co., Ltd., and the estimation accuracy was improved by 38.6% in root mean square error of prediction (RMSEP) compared to the conventional PLS using wavelengths selected on the basis of variable importance on the projection (VIP). The results clearly show that the proposed calibration modeling technique is useful for API content estimation and is superior to the conventional one.     

Combining microwave resonance technology to multivariate data analysis as a novel PAT tool to improve process understanding in fluid bed granulation

A set of 192 fluid bed granulation batches at industrial scale were in-line monitored using microwave resonance technology (MRT) to determine moisture, temperature and density of the granules. Multivariate data analysis techniques such as multiway partial least squares (PLS), multiway principal component analysis (PCA) and multivariate batch control charts were applied onto collected batch data sets. The combination of all these techniques, along with off-line particle size measurements, led to significantly increased process understanding. A seasonality effect could be put into evidence that impacted further processing through its influence on the final granule size. Moreover, it was demonstrated by means of a PLS that a relation between the particle size and the MRT measurements can be quantitatively defined, highlighting a potential ability of the MRT sensor to predict information about the final granule size.This study has contributed to improve a fluid bed granulation process, and the process knowledge obtained shows that the product quality can be built in process design, following Quality by Design (QbD) and Process Analytical Technology (PAT) principles.     

Applications of NIR in early stage formulation development. Part II. Content uniformity evaluation of low dose tablets by principal component analysis

A near infrared method based on principal component analysis (PCA) was developed for predicting content uniformity of low dose tablets manufactured by a direct compression process. The work was conducted in early stage formulation development. NIR spectra of one hundred and eighty tablets from three feasibility batches were used as the pseudo-calibration set. A correlation was established between PCA scores and a set of reference values obtained by HPLC analysis. The reference values were also used to define a concentration range for the active pharmaceutical ingredient to facilitate content uniformity prediction by PCA. Analyses of unknown samples were conducted by forming a prediction set that included the calibration and unknown samples, followed by PCA. Samples from two development batches were predicted using the PCA model and the results were consistent with the reference HPLC values. Remarkably, the model was able to predict CU for tablets that were prepared using different grades of lactose (anhydrous versus monohydrate). Additionally, during this study, the impact of spectrum pretreatments on PCA is demonstrated. A brief discussion is given to highlight the advantages of PCA over partial least squares (PLS) regression for analysis of samples generated in early stage formulation development.     

多变量统计过程控制技术在麦冬提取生产过程的应用研究

目的 建立基于多变量统计过程控制（MSPC）技术的注射用益气复脉（冻干）麦冬水提取过程的在线监测方法,实现对麦冬水提取过程的实时监测。方法 以蒸汽压力、保沸温度、冷却水回水温度3个关键过程参数,结合近红外光谱技术在线监测的果糖水平为变量,采用商业化规模9个生产批次建立麦冬水提取过程的MSPC模型;使用SIMCA-P+14.1软件进行数据分析,使用偏最小二乘算法（PLS）进行自动拟合建立批次变化模型（BEM）,用于生产过程评价;使用主成分分析（PCA）进行自动拟合建立批次水平模型（BLM）,用于批次评价。将模型用于3个商业化规模实验批次（检验批1、2、3）的过程监测,评价模型性能。结果 生成BEM和BLM的HotellingT²图及DMod X控制图,DModX控制图采用＋3SD作为控制限,对各批次参数的数据结构（即各参数的相关关系）进行评价;HotellingT²图以95%作为控制限,在各批次参数的数据结构无差异的情况下,可对各批次数据是否存在异常进行评价。BLM结果显示检验批次的DMod X值超出控制限,BLM结果与BEM检验结果一致,检验批1部分时间节点的DMod X值超出控制限,检验批2和检验批3的大部分时间节点的DModX值超出控制限,对以上超限的数据点进行分析,发现原因主要为冷却水回水温度超出控制水平。结论 借助MSPC技术对复杂中药制造过程进行数据挖掘与模型开发,可实现对中药制药过程的实时监测,为中药智能控制技术的建立提供参考。     

Quality by design approach of a pharmaceutical gel manufacturing process,part 2: Near infrared monitoring of composition and physical parameters

We applied the principles of quality by design to the production process of a pharmaceutical gel by using the near infrared spectroscopy (NIRS) technique in combination with multivariate chemometric tools. For this purpose, we constructed a D‐optimal experimental design having normal operational condition (NOC) batches as central point. The primary aim here was to develop an expeditious NIRS method for determining the composition of a pharmaceutical gel and assess the temporal changes in major physical factors affecting the quality of the product (specifically, viscosity and pH). Gel components were quantified by using partial least squares (PLS) calibration models of the PLS1 type. The study was completed by using the batch statistical process control method to compare product batches included in the experimental design with NOC batches. Similarities and differences between the two types of batches were identified by using control charts for residuals (Q‐statistic) and Hotteling's T2 (D‐statistic). The ensuing models, which were subject to errors less than 5%, allowed the gel production process to be effectively monitored. As shown in this work, the NIRS technique is a highly suitable tool for process analytical technology. © 2011 Wiley‐Liss, Inc. and the American Pharmacists Association J Pharm Sci 100:4442–4451, 2011     

Characterization of a Nonribosomal Peptide Antibiotic Solid Dispersion Formulation by Process Analytical Technologies Sensors

The focus of present investigation was to characterize and evaluate the variability of solid dispersion (SD) of amorphous vancomycin (VCM), utilizing crystalline polyethylene glycol (PEG-6000) as a carrier and subsequently, determining their percentage composition by nondestructive method of process analytical technology (PAT) sensors. The SD were prepared by heat fusion method and characterized for physicochemical and spectral properties. Enhanced dissolution was shown by the SD formulations. Decreased crystallinity of PEG-6000 was observed indicating that the drug was present as solution and dispersed form within the polymer. The SD formulations were homogenous as shown by near infrared (NIR) chemical imaging data. Principal component analysis (PCA) and partial least square (PLS) method were applied to NIR and PXRD (powder X-ray diffraction) data to develop model for quantification of drug and carrier. PLS of both data showed correlation coefficient > 0.9934 with good prediction capability as revealed by smaller value of root mean square and standard error. The model based on NIR and PXRD were two folds more accurate in estimating PEG-6000 than VCM. In conclusion, the drug dissolution from the SD increased by decreasing crystallinity of PEG-6000, and the chemometric models showed usefulness of PAT sensor in estimating percentage of both VCM and PEG-600 simultaneously.     

Batch statistical process control of a fluid bed granulation process using in-line spatial filter velocimetry and product temperature measurements

Fluid bed granulation is a batch process, which is characterized by the processing of raw materials for a predefined period of time, consisting of a fixed spraying phase and a subsequent drying period. The present study shows the multivariate statistical modeling and control of a fluid bed granulation process based on in-line particle size distribution (PSD) measurements (using spatial filter velocimetry) combined with continuous product temperature registration using a partial least squares (PLS) approach. Via the continuous in-line monitoring of the PSD and product temperature during granulation of various reference batches, a statistical batch model was developed allowing the real-time evaluation and acceptance or rejection of future batches.Continuously monitored PSD and product temperature process data of 10 reference batches (X-data) were used to develop a reference batch PLS model, regressing the X-data versus the batch process time (Y-data). Two PLS components captured 98.8% of the variation in the X-data block. Score control charts in which the average batch trajectory and upper and lower control limits are displayed were developed. Next, these control charts were used to monitor 4 new test batches in real-time and to immediately detect any deviations from the expected batch trajectory. By real-time evaluation of new batches using the developed control charts and by computation of contribution plots of deviating process behavior at a certain time point, batch losses or reprocessing can be prevented.Immediately after batch completion, all PSD and product temperature information (i.e., a batch progress fingerprint) was used to estimate some granule properties (density and flowability) at an early stage, which can improve batch release time. Individual PLS models relating the computed scores (X) of the reference PLS model (based on the 10 reference batches) and the density, respectively, flowabililty as Y-matrix, were developed. The scores of the 4 test batches were used to examine the predictive ability of the model.     

Multivariate analysis of nuclear magnetic resonance data--characterization of critical drug substance quality of gentamicin sulfate

Fourteen gentamicin sulfate lots collected from international markets showed high quantities of impurities (30% of studied lots). ¹H NMR spectroscopy as a primary analytical method was applied in order to validate the quantification results obtained from micellar electrokinetic chromatography method (MEKC). In this study, ¹H NMR data of 46 gentamicin sulfate drug substance lots were used to classify the lots by means of principal component analysis (PCA) of 14 ¹H NMR-signals in the 5.0–6.0 ppm region. Three main groups could be classified: high purity (3 lots), average quality (28 lots) and low purity (14 lots); one lot proved to be atypical. The 14 normalized signal heights in the 5.0–6.0 ppm region are predictive for purity quality according to a partial least squares (PLS)-model with sum of all impurities as Y-variable (measured by MEKC).     

Near-Infrared Imaging for Studying Homogeneity of Protein-Sugar Mixtures

Purpose To investigate the applicability of near-infrared (NIR) imaging for assessing the homogeneity of dried protein-sugar formulations.Methods Physical mixtures of lysozyme and trehalose in different ratios were prepared and analyzed by near-infrared (NIR) imaging with a spatial resolution of 10 or 40 μm. To define and select the best imaging strategy, besides visual inspection of the images, several approaches for data processing were tested: single wavelength intensity, peak/height ratio of two specific wavelengths, correlation coefficient with a reference spectrum and principal component analysis (PCA). In order to relate the contrast directly to concentration differences of lysozyme and trehalose, quantitative models were created based on correlation coefficient and partial least squares (PLS) regression. The selected imaging method was applied to compare the homogeneity of a supercritical fluid (SCF) dried and a freeze-dried lysozyme-trehalose mixture.Results All tested methods confirmed each other and showed spatial heterogeneity in the lysozyme and trehalose contents of the physical mixtures. However, multivariate data processing methods (correlation coefficient and PCA/PLS) resulted in more distinct contrasts than univariate approaches (single wavelength analysis) and allowed a quantitative estimation of the homogeneity. As shown by NIR imaging in combination with the correlation coefficient or the PLS method, the SCF dried lysozyme-trehalose formulation was at least as homogeneous as its lyophilized counterpart, at 10 μm pixel size resolution.Conclusions NIR imaging is a useful tool for studying the homogeneity of dried protein-sugar formulations.     

Multivariate data analysis in pharmaceutics: a tutorial review

We provide an overview of latent variable methods used in pharmaceutics and integrated with advanced characterization techniques such as vibrational spectroscopy. The basics of the most common latent variable methods, principal component analysis (PCA), principal component regression (PCR) and partial least-squares (PLS) regression, are presented. Multiple linear regression (MLR) and methods for improved interpretation, variable selection, classification and validation are also briefly discussed. Extensive use of the methods is demonstrated by compilation of the recent literature.     

Determination of Water Penetration and Drug Concentration Profiles in HPMC-Based Matrix Tablets by Near Infrared Chemical Imaging

This work describes methodologies based on near infrared chemical imaging (NIR CI) and chemometric data analysis for studying hydration behaviors of prolonged release tablets, which contain a high solubility drug at high load and a hydrophilic polymer hydro-xypropylmethylcellulose (HPMC). Hydration studies were performed by suspending the tablets in water at ambient temperature. The hydrated tablets were then dissected and scanned by NIR CI. Single wavelength images were obtained for accurately measuring radial dimension of the gel layer and size of the tablet core. By performing a principal component analysis (PCA), the phenomenon of polymer phase transition from the glassy state to rubbery state was detected and visualized. Partial least squares (PLS) models were created for quantitative analysis of the active pharmaceutical ingredient (API) and relative concentration of water in the hydrated tablets. The API concentration profiles are suitable for defining the swelling front in hydrated tablets. Because the NIR CI results are pixel-specific and each pixel has its unique coordinates, it is feasible to analyze and present the results according to spatial locations. The physical and chemical changes at the swelling/diffusion fronts can be demonstrated by overlaying the PCA and PLS results, which shed light on the release mechanism. © 2010 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 99:3081-3088, 2010     

男性高尿酸血症患者血浆代谢组学的研究

目的:研究男性高尿酸血症(HUA)患者血浆成分的变化,为探讨HUA病因、HUA与其他代谢综合征(MS)组分的关系提供参考。方法:运用氢核磁共振(1H-NMR)代谢组学方法分析高尿酸血症患者(HUA组,n=5)和同期体检健康者(健康对照组,n=5)血浆中饱和脂肪酸、N-乙酰糖蛋白、极低密度脂蛋白(VLDL)等代谢物的变化,结合主成分分析(PCA)和偏最小二乘判别分析法(PLS),分析2组这些代谢物差异。结果:HUA组较健康对照组的饱和脂肪酸、N-乙酰糖蛋白、VLDL、乳酸、乙醇的水平上升,卵磷脂、谷氨酰胺、葡萄糖的水平下降。结论:HUA患者较健康对照组存在脂类代谢异常、糖酵解加剧、谷氨酰胺水平下降等问题,可能与HUA的发病、病程转归有关。