Original Article SIMULTANEOUS ANALYSIS OF ISONIAZID AND PYRIDOXINE HYDROCHLORIDE IN TABLET DOSAGE FORM USING PARTIAL LEAST SQUARES BASED ON INFRARED SPECTRA PROFILE LISDA RIMAYANI NASUTION 1* , SEREN ERNEST FLAVIA 3 , YADE METRIPERMATA 1 , MUCHLISYAM BACHRI 1 , HAFID SYAHPUTRA 1 , CHEMAYANTI SURBAKTI 2 1 Department of Pharmaceutical Chemistry, Faculty of Pharmacy, Universitas Sumatera Utara, Medan, Indonesia. 2 Department of Pharmaceutical Biology, Faculty of Pharmacy, Universitas Sumatera Utara, Medan, Indonesia. 3 Undergraduate Program, Faculty of Pharmacy, Universitas Sumatera Utara, Medan, Indonesia * Corresponding author: Lisda Rimayani Nasution; * Email: lisda.rimayani@usu.ac.id Received: 03 Oct 2024, Revised and Accepted: 07 Jan 2025 ABSTRACT Objective: The purpose of this research was to simultaneously analyse a mixture of Isoniazid (INH) and Pyridoxine Hydrochloride (PDX) in tablet dosage form using Partial Least Square (PLS) based on infrared spectra profile. Methods: The calibration models were measured at five concentrations levels (w/w) of INH and PDX within the wavenumber ranges of 779.24- 713.66 cm -1 for INH and 829.39-763.81 cm -1 for PDX. Model performance was evaluated through internal validation using leave-one-out-cross validation (LOO-CV), with accuracy and precision assessed by the coefficient of determination (R 2 ), Prediction Error Sum of Squares (PRESS), and Root mean Square Error of Cross Validation (RMSECV). Divisor approach was applied to separate the spectrum of the individual components within the mixture. Results: The PLS models for INH and PDX levels showed excellent validation results, with R 2 values of 0.9996 and 0.9969, PRESS values of 0.8106 and 0.3129, and RMSCEV values of 0.3675 and 0.2283, respectively. The INH level on INHA ® 400 and Pehadoxin Forte ® tablets were 102.90% and 105.45% of the labelled amount, while the PDX level was 98.93% and 101.57% of the labelled amount, respectively. Conclusion: The PLS model based on infrared spectra profiles demonstrated high accuracy, precision, and effectiveness, offering a non-destructive method for simultaneous analysis of INH and PDX in tablet dosage form. This method is applicable to pharmaceutical quality monitoring. Keywords: Isoniazid, Pyridoxine hydrochloride, Chemometrics, Partial least square © 2025 The Authors. Published by Innovare Academic Sciences Pvt Ltd. This is an open access article under the CC BY license (https://creativecommons.org/licenses/by/4.0/) DOI: https://dx.doi.org/10.22159/ijap.2025v17i2.52853 Journal homepage: https://innovareacademics.in/journals/index.php/ijap INTRODUCTION A reliable quantitative analysis method is essential for determining the concentration of a substance, which is cost-effective, easy to operate, and provides accurate and precise results. Spectrophotometry an analytical method that meets these criteria and is more in demand than other methods because it requires no complex extraction stages, making it more efficient and less time consuming [1, 2]. However, when standard spectrophotometric methods are used to measure the amounts of complex mixtures at the same time, problems arise because the spectra overlaps, making it harder to accurately measure each component [3]. Therefore, the development of a quantifiable analytical method is required to address this problem. The development of Fourier Transformed Infrared (FTIR) spectrophotometry has proven to be a viable solution for the quantitative analysis of complex mixtures [4]. FTIR provide numerous benefits for quantitative analysis, such as its rapid nature, non- destructive approach, simple to prepare, and low solvent needs. Moreover, FTIR is gaining recognition as a sustainable alternative, as it reduces environmental issues linked to solvent usage [5]. In its application, FTIR can be used in conjunction with chemometrics techniques, particularly Partial Least Square (PLS) to enhance its analytical capabilities. Chemometrics is an interdisciplinary field that applies mathematical and statistical techniques to process and analyse multivariate data [6]. The main advantages of chemometrics, particularly PLS, is its capacity to handle overlapping spectra data and predicted component concentration without requiring separation processes. Furthermore, PLS techniques offer ease of use, cost-effectiveness, and higher sensitivity compared to other techniques [7]. In a previous study, the combination of FTIR and PLS provided advantages in terms of high accuracy and precision for the separation multicomponent mixtures, with the calibration model processing only the desired analyte concentrations [8]. However, one limitation of multivariate calibration methods is the potential occurrence of overfitting. Therefore, validation performed using leave-one-out-a cross-validation (LOO-CV) for minimize overfitting also obtain accurate, specific, and reliable results [9]. While the PLS model demonstrates strong predictive capabilities, it does exhibit certain limitations when applied to samples with varying matrix compositions, particularly those containing new excipients that can lead to increased prediction errors [10]. To address this, it is essential to consider strategies such as model optimization or calibration adjustments to improve the accuracy. One such strategy involves using individual calibration curves, as suggested by Pinto et al. [11]. Advancements in scientific knowledge have led to the growing availability of pharmaceutical products. Currently, many pharmaceutical formulations contain multiple active ingredients [12]. A specific example is the combination of isoniazid (INH) and pyridoxine hydrochloride (PDX), which poses challenges in quality control due to the similarities in their physicochemical properties and overlapping absorbance spectra [13]. In previous studies, various techniques have been employed for the quantification of INH and PDX in tablet dosage forms, such as Reverse Phase-High Performance Liquid Chromatography (RP-HPLC) [14]. However, RP-HPLC requires complex sample preparation, costly equipment, and specialized expertise, which limits its widespread application [14]. Other methods, such as zero-crossing derivative ultraviolet (UV) spectrophotometry, offer a simpler approach with less solvent consumption, but involve lengthy derivation processes to ensure reliable results [15]. Similarly, the absorbance ratio UV spectrophotometry method, which relies on iso absorbance points, has demonstrated limitations in specificity and sensitivity, making it less suitable for accurate quantitative analysis [16, 17]. Notably, there has been no prior study utilizing the FTIR-PLS method for the quantification INH and PDX contents. Given these limitations, the present research aims to develop a more efficient and accurate method for the quality control of INH and PDX in International Journal of Applied Pharmaceutics ISSN- 0975-7058 Vol 17, Issue 2, 2025