Evaluation of key sources of variability in the measurement of pharmaceutical drug products by near infrared reflectance spectroscopy

doi:10.1016/S0731-7085(97)00229-X

Journal of Pharmaceutical and Biomedical Analysis

Volume 17, Issues 4–5, August 1998, Pages 641-650

https://doi.org/10.1016/S0731-7085(97)00229-X Get rights and content

Abstract

Potential sources of variability in the measurement of solid oral drug products by near infrared reflectance spectroscopy were evaluated with statistical experimental design. Spectra were collected for two different tablet types according to the data collection and treatment parameters defined by the experimental design. Each tablet had three different dose-levels. Libraries were constructed using second-derivative spectra. Key figures-of-merit generated during internal and external library validation were used to calculate which parameters most strongly influence the library performance for dose-level discrimination. These responses and their corresponding experimental conditions were evaluated with the screening model in the JMP® program. Segment value used for the second-derivative calculation was an influential factor and had a complex effect. Orientation on the sampling platform also had an influential effect for embossed tablets. Collection of spectra over fewer days decreased variability within the library. More frequent reference spectrum collection improved the performance of libraries to a small degree. A larger sample population increased the range of spectral variability within a dose-level but apparently not the overall performance of the library. The number of scans averaged per spectrum was not an influential factor in this study. These results are summarized and used to recommend an approach to dose-level discrimination.

Introduction

Near infrared reflectance spectroscopy (NIRS) is a rapid analysis technique that has grown in its use in the pharmaceutical industry 1, 2, 3. Applications include water determination 4, 5, identification 6, 7, 8, 9, evaluation of mixing homogeneity [10], and quantitative quality control 11, 12. In all cases, the most important aspect of a successful measurement is the construction of an appropriate calibration set and the control or inclusion of all significant sources of variability.

A primary goal in the authors' laboratory has been to understand the parameters that affect the ability to use NIRS to differentiate between dose-levels of drug products. As applied to final package identification, this testing is one aspect of clinical material analysis in the pharmaceutical industry. Feasibility studies have appeared in the literature 7, 8but little work has been presented on the effect of data collection and treatment parameters on this type of measurement. The work presented here provides a scientifically based model upon which the authors can recommend operating conditions for good sensitivity and efficiency. In daily operation, analysts must discriminate between low and closely spaced dose-levels, making this information valuable.

In this study, statistical experimental designs [13]were used to evaluate how selected sources of variability impact the quality of a spectral library built for dose-level discrimination. Parameters included instrument settings such as the number of scans averaged per spectrum, data treatment settings such as the segment used for second-derivative calculation, and the design of the library such as the number of dose units scanned. Plackett–Burman designs were generated that defined a series of experiments using different settings of these parameters. Responses from each experiment were chosen that correlated to the ability of the library to separate between dose-levels. In general terms, a better library is one where the spectra for a given dosage level are tightly grouped and the separation between the mean spectra of different dosage levels is the largest. Responses were chosen to highlight these traits. This approach allowed the evaluation of many parameters under actual conditions of the desired measurement in a way that was not possible when doing experiments that only vary one parameter at a time.

Section snippets

Materials and instrumentation

Typical tablets, A and B, under development at Eli Lilly and Company were used for this study. There were three dosage levels of tablet A (25, 50, and 75 mg drug per 200-mg tablet) and three dosage levels of tablet B (30, 60, and 150 mg drug per 250-mg tablet). Both tablets were oblong and tablet A was embossed on both sides. Tablet B was smooth on both sides. Three hundred tablets of each type were used.

Diffuse reflectance spectra of these tablets were obtained with an NIRSystems 6500

Spectral regions

Proper dose-level discrimination depends on the selection of spectral region that corresponds to absorption bands of the active drug substance. In the NIR region, most absorption bands arise from the overtone and combination bands of the fundamental infrared vibrations of C–H, N–H, and O–H bonds.

Significant changes of NIR spectral features of these tablets have been observed when dose-level varies. For example, as the active ingredient of tablet B increased, the absorption band of the

Conclusion

This work has shown the influence and significance of various factors on the ability of NIRS to discriminate between dose-levels of tablets. Segment value was a significant factor with good results achieved by using a segment value of 10 when the libraries were built from spectra collected over several days. Orientation was a significant factor only for embossed tablets. Total number of samples was a major factor when focusing the response on uniformity between doses (QDFMS) but was not a

Acknowledgements

The authors would like to thank Dr Bobby Snider and Ms Maryanne Wagner for their review of the manuscript.

References (13)

I.R. Last et al.
J. Pharm. Biomed. Anal.
(1993)
M.A. Dempster et al.
J. Pharm. Biomed. Anal.
(1993)
F. Gonzalez et al.
J. Pharm. Biomed. Anal.
(1995)
W. Plugge et al.
J. Pharm. Biomed. Anal.
(1993)
M. Blanco et al.
Talanta
(1993)
E.W. Ciurczak
Appl. Spectrosc. Rev.
(1987)

There are more references available in the full text version of this article.

Cited by (19)

Tablet splitting: Product quality assessment of metoprolol succinate extended release tablets
2010, International Journal of Pharmaceutics
Citation Excerpt :
For this purpose, approximately 250 mg of each raw material was compressed into 10 mm diameter tablets on a hydraulic tablet press (Carver Inc.) using 10 kN pressure for 10 s. Prior to data analysis, all raw reflectance data (R) were transformed into absorbance (A) by the relation A = −log10(1/R). The raw NIR diffuse reflectance spectra obtained from a NIR chemical imaging measurement contain both chemical and non-chemical information about the solid sample (Borer et al., 1998). The source of the non-chemical information may be from the sample (e.g. uneven sample surface or differences in sample density) and/or the instrumentation (e.g. changes in lamp intensity or detector response).
Metoprolol succinate extended release tablets comprise a multiple unit system containing metoprolol succinate in a multitude of controlled release pellets. Each pellet acts as a separate drug delivery unit and is designed to deliver metoprolol continuously over the dosage interval. Despite the flexibility that controlled release pellets may offer, segregation is one of the challenges that commonly occur during tableting for such drug delivery system. Since all commercial metoprolol succinate extended release tablets are scored, they are deemed suitable for splitting. The present study was aimed at utilizing an innovative technology to determine the dose uniformity for split tablets. Four marketed drug products consisting of innovator and generics were evaluated for effect of splitting on weight, assay and content uniformity. Novel analytical tool such as near infrared (NIR) chemical imaging was used to visualize the distribution of metoprolol succinate and functional excipients on the surfaces of the marketed tablets. The non-homogeneous distribution of directly compressed metoprolol succinate beads on the surface of the tablets as well as the split intersection explained the large variation in the split tablets’ weight and content uniformity results. The obtained results indicated the usefulness of NIR chemical imaging to determine the need for content uniformity studies for certain split tablets.
Near-infrared chemical imaging (NIR-CI) on pharmaceutical solid dosage forms-Comparing common calibration approaches
2008, Journal of Pharmaceutical and Biomedical Analysis
Near-infrared chemical imaging (NIR-CI) is the fusion of near-infrared spectroscopy and image analysis. It can be used to visualize the spatial distribution of the chemical compounds in a sample (providing a chemical image). Each sample measurement generates a hyperspectral data cube containing thousands of spectra. An important part of a NIR-CI analysis is the data processing of the hyperspectral data cube. The aim of this study was to compare the ability of different commonly used calibration methods to generate accurate chemical images. Three common calibration approaches were compared: (1) using single wavenumber, (2) using classical least squares regression (CLS) and (3) using partial least squares regression (PLS1). Each method was evaluated using two different preprocessing methods.
A calibration data set of tablets with five constituents was used for analysis. Chemical images of the active pharmaceutical ingredient (API) and the two major excipients cellulose and lactose in the formulation were made. The accuracy of the generated chemical images was evaluated by the concentration prediction ability. The most accurate predictions for all three compounds were generated by PLS1. The drawback of PLS1 is that it requires a calibration data set and CLS, which does not require a calibration data set, therefore proved to be an excellent alternative. CLS also generated accurate predictions and only requires the pure compound spectrum of each constituent in the sample. All three calibration approaches were found applicable for hyperspectral image analysis but their relevance of use depends on the purpose of analysis and type of data set. As expected, the single wavenumber method was primarily found useful for compounds with a distinct spectral band that was not overlapped by bands of other constituents.
This paper also provides guidance for hyperspectral image (or NIR-CI) analysis describing each of the typical steps involved.
Tablet identification using near-infrared spectroscopy (NIRS) for pharmaceutical quality control
2008, Journal of Pharmaceutical and Biomedical Analysis
A need for more reliable and faster analytical methods for the identification of the active pharmaceutical ingredient (API) in finished pharmaceutical products is launched by the International Conference on Harmonisation (ICH) of Technical Requirements for Registration of Pharmaceuticals for Human Use, Test Procedures and Acceptance Criteria for New Drug Substances and New Drug Products: Chemical Substances, Q6A (1999). The use of infrared spectroscopy is suggested as a means to obtain specific identification. Near-infrared spectroscopy (NIRS) is a reliable method that offers important advantages for the large-scale production of tablets, such as high-throughput and accurate multiparametric data collection. Despite the grown number of reported NIRS identification methods, only a few methods have been approved by the regulatory authorities, which might be due to difficulties on clearly presenting the methods in official documents and audits. Motivated by the lack of clear protocols for the NIRS method's development, here we propose a process for building reliable identification NIRS methods.
For illustration purposes, a method is described for the identification of API in coated tablets containing 2%, 4% and 8% of thiamazole. The method described was successfully validated according to the International Conference on Harmonisation (ICH) of Technical Requirement for Registration of Pharmaceuticals for Human Use, Validation of Analytical Procedures: Text and Methodology, Q2 (2005). The described method was subsequently approved by European national authorities and thus is suitable for use in cGMP environment.
Near-infrared spectroscopy applications in pharmaceutical analysis
2007, Talanta
Citation Excerpt :
The active compound in granulates can also be assayed by means of NIR spectroscopy [69]. Many applications deal with the quantification of a substance in tablets or tablet cores [50,59,67,68,70,110,142,159,180–198]. The majority uses PLS or MLR methods to relate the spectral information to the concentration of the compound.
Near-infrared (NIR) spectroscopy is a fast and non-destructive analytical technique that offers many advantages for a broad range of industrial applications. In this work, we reviewed recent developments in the pharmaceutical domain where it can be applied from raw material identification to final product release. The characteristics of NIR allow the technique to be implemented as a process analytical technology (PAT). Moreover, recent instrumental developments open the perspectives of numerous applications in the NIR imaging area. After “Introduction”, according to their subject, the applications are discussed in the parts “Identification”, “Water content”, “Assay” and “Other applications”.
Near infrared spectroscopy and process analytical technology to master the process of busulfan paediatric capsules in a university hospital
2006, Journal of Pharmaceutical and Biomedical Analysis
The prescription of unlicensed oral medicines in paediatrics leads the hospital pharmacists to compound hard capsules, such as busulfan, an alkylating agent prescribed in preparative regimens for bone marrow transplantation. In this study, we have investigated how the general principle of process analytical technology (PAT) can be implemented at the small size of our hospital pharmacy manufacturing unit.
Near infrared spectroscopy (NIRS) was calibrated for raw material identification, blend uniformity analysis and final content uniformity of busulfan hard capsules of 11 different strengths. Measurements were performed on capsules from 2 to 40 mg (n = 440). After optimisation, accuracy and linearity of the NIRS quantitative method was demonstrated after comparison with a previously validated quantitative high performance thin layer chromatography (HPTLC) method. Such a comparison led to attractive NIRS precision: ±0.7 to ±1.0 mg for capsules from 2 to 40 mg, respectively.
As NIRS is a rapid and non-destructive technique, the individual control of a whole batch of busulfan paediatric capsules intended to be administrated is possible. Actually, mastering the process of busulfan paediatric capsules with the NIRS integrated into the notion of PAT is a powerful analytical tool to assess the process quality and to perform content uniformity of at least 5 mg busulfan-containing capsules.
The effect of preprocessing methods in reducing interfering variability from near-infrared measurements of creams
2004, Journal of Pharmaceutical and Biomedical Analysis
This work is part of a study in which the possibility of NIR combined with some chemometrical methods is investigated as a suitable technique to classify clinical study samples of a cream. In this study, the influence of different preprocessing methods on the removal of spectral variations due to some variance sources has been investigated. The applied preprocessing methods are standard normal variate (SNV), detrend correction, offset correction, and first and second derivation. The investigated variance sources are different batches of ingredients, different samples of the same batch, different days and different positions of the sample cup in the sample drawer of the instrument. A nested ANOVA design has been applied in order to quantify the variances introduced by these variance sources. Since ANOVA is a univariate technique, the necessary variable (wavelength) selection has been performed by the Fisher criterion. The best results, i.e. largest reduction of interfering variability and clearest distinction between different clinical study samples, are obtained with the second derivative spectra.

View all citing articles on Scopus

View full text

Evaluation of key sources of variability in the measurement of pharmaceutical drug products by near infrared reflectance spectroscopy

Abstract

Introduction

Section snippets

Materials and instrumentation

Spectral regions

Conclusion

Acknowledgements

J. Pharm. Biomed. Anal.

J. Pharm. Biomed. Anal.

J. Pharm. Biomed. Anal.

J. Pharm. Biomed. Anal.

Talanta

Appl. Spectrosc. Rev.