Optimization of wavelength range and data interval in chemometric analysis of complex pharmaceutical mixtures

Rapid screening and multicomponent quantifications of active components of oral syrup over-the-counter medications by Raman and UV-visible spectroscopy and multivariate regression analysis

Over-the-counter medications (OTCMs) are frequently recommended as a first-line treatment for common ailments, diseases, and illnesses. Oral liquid dosage forms are advantageous for rapid absorption with no dissolution time and are easier for pediatric and geriatric consumers to swallow. The production of these medicines by pharmaceutical industry makes them readily available to the public. Although the US Food and Drug Administration (FDA) provides strict guidelines to drug manufacturers of these products; the risk of counterfeiting is a global issue. This can lead to several adverse effects and health issues. Here, we report a fast screening and quality assurance method using Raman and UV-visible spectroscopy combined with Principal Component Analysis (PCA) and Partial-Least-Square (PLS) regression of commonly used OTCM oral syrups. PLS regressions of UV-visible absorption spectra were used for multicomponent quantifications of the active component (acetaminophen, guaifenesin, dextromethorphan HBr, and phenylephrine HCl) concentrations of OTMCs in flavored (sugar or sugar-free) oral syrups. Raman and UV-visible spectral responses varied based on the type and concentration of the active component analyzed. PCA of the spectral data provided pattern recognition of the oral syrup OTCM. The developed PLS method demonstrated good linearity with an R2 > 0.9784 and high sensitivity with a low detection limit of 0.02 mg/mL for acetaminophen and guaifenesin. Moreover, the simultaneous quantification of concentrations of all active components by the described method yielded good accuracies ranging from 88 to 94%. This study provides an example of the benefits of the combined use of Raman and UV-vis spectral profiling, PCA, and PLS regression for the quality analysis of oral syrups OTCM providing multicomponent quantification of active components with no need for sample extraction. The reported method can be easily adapted and scaled for online detection analysis used in the drug manufacturing industry, both in-situ and field analysis, and for the quality control of syrups OTCM by regulatory agencies and quality control officers.

Quantification of ciprofloxacin in pharmaceutical products from various brands using FT-NIR: A comparative investigation of PLS and MCR-ALS

This study aims to quantify ciprofloxacin in commercial tablets with varying excipient compositions using Fourier Transform Near-Infrared Spectroscopy (FT-NIR) and chemometric models: Partial Least Squares (PLS) and Multivariate Curve Resolution - Alternating Least Squares (MCR-ALS). Matrix variation, arising from differences in excipient compositions among the tablets, can impact quantification accuracy. We discuss this phenomenon, emphasizing potential issues introduced by varying certain excipients and its importance in reliable ciprofloxacin quantification. We evaluated the performance of PLS and MCR-ALS models independently on two sets of tablets, each containing the same drug substance but different excipients. The statistical results revealed promising results with PLS prediction error of 0.38% w/w of the first set and 0.47% w/w of the second set, while MCR-ALS achieved prediction errors of 0.67% w/w of the first set and 1.76% w/w of the second set. To address the challenge of matrix variation, we developed single models for PLS and MCR-ALS using a dataset combining both first and second sets. The PLS single model demonstrated a prediction error of 4.3% w/w and a relative error of 6.41% w/w, while the MCR-ALS single model showed a prediction error of 1.88% w/w and a relative error of 1.29% w/w. We then assessed the performance of the single PLS and MCR-ALS models developed based on the combination of the first and the second set in quantifying ciprofloxacin in various commercial tablet brands containing new excipients. The PLS model achieved a prediction error ranging between 6.2% w/w and 8.39% w/w, with relative errors varied between 8.53% w/w and 12.82% w/w. On the other hand, the MCR-ALS model had a prediction error between 1.11% w/w and 2.66% w/w, and the relative errors ranging from 0.8% to 1.74% w/w.

QbD-steered development and validation of an RP-HPLC method for quantification of ferulic acid: Rational application of chemometric tools

The present work describes the systematic development of a simple, rapid, sensitive, robust, effective and cost-effective reversed-phase high performance liquid chromatographic method for quantitative analysis of ferulic acid using analytical quality by design paradigms. Initially, apt wavelength for the analysis of ferulic acid was selected employing principal component analysis as the chemometric tool. An Ishikawa fishbone diagram was constructed to delineate various plausible variables influencing analytical target profile, viz. peak area, theoretical plate count, retention time and peak tailing as the critical analytical attributes. Risk assessment using risk estimation matrix and factor screening studies employing Taguchi design aided in demarcating two critical method parameters, viz. mobile phase ratio and flow rate affecting critical analytical attributes. Subsequently, the optimum operational conditions of the liquid chromatographic method were delineated using face-centred composite design. Multicollinearity among the chosen factors for optimization was analyzed by the magnitude of variance inflation factor optimized analytical design space, providing optimum method performance, was earmarked using numerical and graphical optimization and corroborated using Monte Carlo simulations. Validation, as per the ICH Q2(R1) guidelines, ratified the efficiency and sensitivity of the developed novel analytical method of ferulic acid in the mobile phase and the human plasma matrix. The optimal method used a mobile phase, comprising of acetonitrile: water (47:53% v/v, pH adjusted to 3.0 with glacial acetic acid), at a flow rate of 0.8 mL·min^-1, at a λ_max of 322 nm using a C₁₈ column. Use of principal component analysis unearthed the suitable wavelength for analysis, while analytical quality by design approach, along with Monte Carlo simulations, facilitated the identification of influential variables in obtaining the "best plausible" validated chromatographic solution for efficient quantification of ferulic acid.

A UV/Vis Spectroscopy-Based Assay for Monitoring of Transformations Between Nucleosides and Nucleobases

Efficient reaction monitoring is crucial for data acquisition in kinetic and mechanistic studies. However, for conversions of nucleosides to their corresponding nucleobases, as observed in enzymatically catalyzed nucleoside phosphorylation reactions, the current analytical arsenal does not meet modern requirements regarding cost, speed of analysis and high throughput. Herein, we present a UV/Vis spectroscopy-based assay employing an algorithm for spectral unmixing in a 96-well plate format. The algorithm relies on fitting of reference spectra of nucleosides and their bases to experimental spectra and allows determination of nucleoside/nucleobase ratios in solution with high precision. The experimental procedure includes appropriate dilution of a sample into aqueous alkaline solution, transfer to a multi-well plate, measurement of a UV/Vis spectrum and subsequent in silico spectral unmixing. This enables data collection in a high-throughput fashion and reduces costs compared to state-of-the-art HPLC analyses by approximately 5-fold while being 20-fold faster and offering comparable precision. Additionally, the method is robust regarding dilution and sample transfer errors as it only considers spectral form and not absolute intensity. It can be applied to all natural nucleosides and nucleobases and even unnatural ones as demonstrated by several examples.

Evaluation of human breastmilk adulteration by combining Fourier transform infrared spectroscopy and partial least square modeling

A two-step chemometric procedure was developed on the attenuated total reflection-Fourier transform infrared data of human breastmilk to detect adulteration by water or cow milk. The samples, collected from a Milk Bank, were analyzed before and after adulteration with whole, skimmed, semi-skimmed cow milk and water. A preliminary clustering via principal component analysis distinguished three classes: pure milk, milk adulterated with water, and milk adulterated with cow milk. A first partial least square-discriminant analysis (PLS-DA) classification model was built and then applied on new samples to identify the specific adulterants. The external validation on this model reached 100% of the correct identification of pure milk and 90% of the type of adulterants. In the following step, four PLS calibration models were built to quantify the amount of the adulterant detected in the classification analysis. The prediction performance of these models on new samples showed satisfactory parameters with root mean square error of prediction and percentage relative error lower than 1.38% and 3.31%, respectively.

Predictive modelling of colossal ATR-FTIR spectral data using PLS-DA: empirical differences between PLS1-DA and PLS2-DA algorithms

In response to our review paper [L. C. Lee et al., Analyst, 2018, 143, 3526-3539], we present a study that compares empirical differences between PLS1-DA and PLS2-DA algorithms in modelling a colossal ATR-FTIR spectral dataset. Over the past two decades, partial least squares-discriminant analysis (PLS-DA) has gained wide acceptance and huge popularity in the field of applied research, partly due to its dimensionality reduction capability and ability to handle multicollinear and correlated variables. To solve a K-class problem (K > 2) using PLS-DA and high-dimensional data like infrared spectra, one can construct either K one-versus-all PLS1-DA models or only one PLS2-DA model. The aim of this work is to explore empirical differences between the two PLS-DA algorithms in modeling a colossal ATR-FTIR spectral dataset. The practical task is to build a prediction model using the imbalanced, high dimensional, colossal and multi-class ATR-FTIR spectra of blue gel pen inks. Four different sub-datasets were prepared from the principal dataset by considering the raw and asymmetric least squares (AsLS) preprocessed forms: (a) Raw-global region; (b) Raw-local region; (c) AsLS-global region; and (d) AsLS-local region. A series of 50 models which includes the first 50 PLS components incrementally was constructed repeatedly using the four sub-datasets. Each model was evaluated using six different variants of v-fold cross validation, autoprediction and external testing methods. As a result, each PLS-DA algorithm was represented by a number of figures of merit. The differences between PLS1-DA and PLS2-DA algorithms were assessed using hypothesis tests with respect to model accuracy, stability and fitting. On the other hand, confusion matrices of the two PLS-DA algorithms were inspected carefully for assessment of model parsimony. Overall, both the algorithms presented satisfactory model accuracy and stability. Nonetheless, PLS1-DA models showed significantly higher accuracy rates than PLS2-DA models, whereas PLS2-DA models seem to be much more stable compared to PLS1-DA models. Eventually, PLS2-DA also proved to be less prone to overfitting and is more parsimonious than PLS1-DA. In conclusion, the relatively high accuracy of the PLS1-DA algorithm is achieved at the cost of rather low parsimony and stability, and with an increased risk of overfitting.

Investigation of Drug-Polymer Compatibility Using Chemometric-Assisted UV-Spectrophotometry

A simple chemometric-assisted UV-spectrophotometric method was used to study the compatibility of clindamycin hydrochloride (HC1) with two commonly used natural controlled-release polymers, alginate (Ag) and chitosan (Ch). Standard mixtures containing 1:1, 1:2, and 1:0.5 w/w drug-polymer ratios were prepared and UV scanned. A calibration model was developed with partial least square (PLS) regression analysis for each polymer separately. Then, test mixtures containing 1:1 w/w drug-polymer ratios with different sets of drug concentrations were prepared. These were UV scanned initially and after three and seven days of storage at 25 °C. Using the calibration model, the drug recovery percent was estimated and a decrease in concentration of 10% or more from initial concentration was considered to indicate instability. PLS models with PC3 (for Ag) and PC2 (for Ch) showed a good correlation between actual and found values with root mean square error of cross validation (RMSECV) of 0.00284 and 0.01228, and calibration coefficient (R²) values of 0.996 and 0.942, respectively. The average drug recovery percent after three and seven days was 98.1 ± 2.9 and 95.4 ± 4.0 (for Ag), and 97.3 ± 2.1 and 91.4 ± 3.8 (for Ch), which suggests more drug compatibility with an Ag than a Ch polymer. Conventional techniques including DSC, XRD, FTIR, and in vitro minimum inhibitory concentration (MIC) for (1:1) drug-polymer mixtures were also performed to confirm clindamycin compatibility with Ag and Ch polymers.

Development and validation of multivariate calibration methods for simultaneous estimation of Paracetamol, Enalapril maleate and hydrochlorothiazide in pharmaceutical dosage form

Three multivariate calibration spectrophotometric methods were developed for simultaneous estimation of Paracetamol (PARA), Enalapril maleate (ENM) and Hydrochlorothiazide (HCTZ) in tablet dosage form; namely multi-linear regression calibration (MLRC), trilinear regression calibration method (TLRC) and classical least square (CLS) method. The selectivity of the proposed methods were studied by analyzing the laboratory prepared ternary mixture and successfully applied in their combined dosage form. The proposed methods were validated as per ICH guidelines and good accuracy; precision and specificity were confirmed within the concentration range of 5-35μgmL^-1, 5-40μgmL^-1 and 5-40μgmL^-1of PARA, HCTZ and ENM, respectively. The results were statistically compared with reported HPLC method. Thus, the proposed methods can be effectively useful for the routine quality control analysis of these drugs in commercial tablet dosage form.