Quantification of Amlodipine Maleate Content in Amorphous Solid Dispersions Produced by Fluidized Bed Granulation Using Process Analytical Technology Tools

Abstract

Background: Active pharmaceutical ingredient (API) content is a critical quality attribute (CQA) of amorphous solid dispersions (ASDs) prepared by spraying a solution of APIs and polymers onto the excipients in fluid bed granulator. This study presents four methods for quantifying API content during ASD preparation.

Methods: Raman and three near-infrared (NIR) process analysers were utilized to develop methods for API quantification. Four partial least squares (PLS) models were developed using measurements from three granulation batches, with an additional batch used to evaluate model predictability. Models performance was assessed using metrics such as root mean square error of prediction (RMSEP), root mean square error of cross-validation (RMSECV), residual prediction deviation (RPD), and others.

Results: Off-line and at-line NIR models were identified as suitable for process control applications. Additionally, at-line Raman measurements effectively predicted the endpoint of the spraying phase.

Conclusions: To the best of authors’ knowledge, this is the first study focused on monitoring API content during fluidized bed granulation (FBG) used for ASD preparation. The findings provide novel insights into the application of Raman and NIR process analysers with PLS modelling for monitoring and controlling ASD preparation processes.

Introduction

A large percentage of marketed drugs (~40%), as well as those in development (~90%), exhibit poor aqueous solubility [1]. Drug candidates with low solubility present challenges, such as reduced bioavailability and a higher likelihood of being removed from the development process [2]. The preparation of amorphous solid dispersions (ASDs) is a formulation strategy used for improving the bioavailability of poorly soluble active pharmaceutical ingredients (APIs). This approach involves dispersing APIs in a polymer matrix. ASD formulations improve the bioavailability of the APIs by enhancing its apparent solubility and dissolution rate. These improvements result from the higher free energy, along with the increased thermodynamic and chemical activity, of amorphous APIs compared to their crystalline counterparts [3].

The preparation of an ASD using fluidized bed granulation (FBG) is categorised as a solvent-based method, where the API is first dissolved in an organic or aqueous solvent, then sprayed onto the fluidized excipients, and simultaneously dried. During this process, ASD is layered onto the excipient cores [1,4,5]. One of the most important critical quality attributes (CQAs) of the granules produced by spraying the API solution is the API content. Various factors can cause deviations from the expected API content. For instance, granulation liquid may be sprayed onto the chamber walls, or droplets may be spray-dried and carried through the filters, leading to lower API content. Conversely, the API content may be higher than expected due to selective excipient loss through the distribution plate or filters. Therefore, achieving the target API content can be challenging when the only parameter used to estimate API content is the amount of granulation liquid sprayed.

In 2002, the United States Food and Drug Administration (FDA) addressed the issue of pharmaceutical processes being monitored and controlled solely by focusing on process parameters, without real-time insight into the status of CQAs. The initiative, titled “Pharmaceutical Current Good Manufacturing Practices (CGMPs) for the 21st Century”, aimed to ensure quality is built into the product throughout the manufacturing process [6]. The key tool of the initiative was process analytical technology (PAT), defined as a system for designing, analysing, and controlling manufacturing through the timely measurements of critical performance attributes (CPAs) and CQAs of processes and in-process materials. Measurements are performed using process analysers, which capture the data in real or near real time. The measurements can be performed: in-line (the sample is measured during processing), on-line (the sample is removed from the process, measured, and possibly returned), and at-line (the sample is removed from the process and analysed nearby). Analysis conducted in a laboratory setting is referred to as off-line analysis [7].

Both near-infrared (NIR) spectroscopy and Raman spectroscopy are commonly used PAT analysers that allow fast, cost-effective, and non-destructive measurement collection. The resulting spectra provide information on both the physical and chemical properties of the analysed materials [8,9].

There is an existing body of literature on the application of NIR spectroscopy for quantification of the API content during FBG. Gavan et al. used an NIR analyser for in-line monitoring of total polyphenolic content (TPC) during the process of spraying A. genevensis liquid extract onto a mixture of lactose and microcrystalline cellulose (MCC) in a fluidized bed (FB) processing chamber. An orthogonal partial least squares (OPLS) model was developed to predict TPC from the NIR spectra, achieving an R2 of 0.982 for correlated variability and a Q2 of 0.951. The root mean square error of cross-validation (RMSECV) was 31.5 mg gallic acid per 100 g dry extract, while the finished dry extract contained 469.25 mg of gallic acid per 100 g dry extract. The authors noted that the developed method detected the process endpoint when the desired TPC concentration was reached [10]. Roggo et al. applied an in-line NIR probe to monitor the API content in the granulate exiting the FB processing chamber and in the tablet press feed frame. The specific API used was not disclosed. A single partial least squares (PLS) model was developed, achieving R2 of 0.996 and RMSECV of 1.25% [11]. Zhao et al. developed multiple models for monitoring the content of three APIs (albiflorin, paeoniflorin, and benzoylpaeoniflorin) from NIR spectra collected off-line from granulate samples. These three APIs were present in the herbal extract, which was sprayed onto maltodextrin during the FBG process. PLS, the genetic algorithm interval PLS (GA-iPLS), back-propagation artificial neural network (BP-ANN), and particle swarm optimisation support vector machine (PSO-SVM) models were developed, with the best-performing PSO-SVM models achieving R2 values between 0.978 and 0.985, and root mean square error of prediction (RMSEP) values of 0.0947 (albiflorin), 0.2850 (paeoniflorin), and 0.0134 (benzoylpaeoniflorin) [12]. Zhong et al. applied in-line NIR spectrometry to monitor nifedipine content uniformity during FBG, where a granulation liquid containing a binder was sprayed onto a mixture of nifedipine, MCC, and lactose. PLS and extended iterative optimisation technique (EIOT) models were developed, achieving R2 values of 0.982 and 0.968, Q2 values of 0.612 and 0.885, and RMSEP values of 3.40% and 2.77%, respectively [13].

To the best of the authors’ knowledge, no studies have focused on the quantification of API content during FBG using Raman spectroscopy. However, there are studies that have applied Raman spectroscopy for API quantification during other pharmaceutical processes. Harting and Kleinebudde applied in-line Raman spectroscopy to twin-screw wet granulation for the quantification of ibuprofen and diclofenac sodium. Separate PLS models were developed for each API. The ibuprofen model achieved a Q2 of 0.993, and an RMSEP of 0.59%, while the diclofenac sodium model achieved a Q2 of 0.997, and an RMSEP of 1.50% [14]. In a subsequent study, they optimised the in-line Raman setup by including interior lighting. They developed “dark” and “light” PLS models for diclofenac sodium prediction, with R2 values of 0.975 and 0.923, Q2 of 0.998 for both models, and RMSEP of 0.31% and 0.29%, respectively [15]. Müller et al. applied in-line Raman spectroscopy to monitor diprophylline content in the film coat during the tablet film coating process in a pan coater. In this process, a dispersion containing diprophylline was used for the tablet coating. Multiple PLS models were developed to predict relative (RMSEP: 0.6322%) weight gain, diprophylline content increase (RMSEP: 3.361%), and total diprophylline content (RMSEP: 0.4258 mg) in the coated tablets [16]. In subsequent studies, the monitoring method was validated according to ICH guideline Q2 and transferred from a mini scale pan coater to a micro-scale coater [17,18].

The aim of this study was to develop methods for the quantification of amlodipine maleate (AM) during the FBG process used for producing an ASD. This was achieved by using NIR and Raman PAT process analysers, capable of performing in-line and at-line measurements. The developed methods could be used for process end-point detection and real-time product release.

This article distinguishes itself from the previous research by focusing on the FBG process used to produce granulate containing an ASD, where an API and polymer solution is sprayed onto an inert powder. In contrast, earlier studies using NIR spectroscopy have mainly focused on monitoring API content during the spraying of API liquid extracts without polymers onto inert powders in a FB processing chamber [10,12], evaluating the API content in the finished granulate as it exits the chamber [11], or assessing the API content uniformity when APIs were incorporated directly into the initial powder mixture rather than being sprayed [13]. Additionally, earlier studies using Raman spectroscopy have monitored API content during twin-screw wet granulation [14,15], and tablet film coating with dispersions containing APIs [17,18], with none focusing on FBG. This article represents the first use of both Raman and NIR spectroscopy PAT analysers for determining API content within ASDs produced by FBG, as well as the first combined application of Raman and NIR spectroscopy for API content prediction during the FBG process.

Download the full article as PDF here Quantification of Amlodipine Maleate Content in Amorphous Solid Dispersions Produced by Fluidized Bed Granulation Using Process Analytical Technology Tools

or read it here

Materials

Amlodipine maleate (AM) was supplied by Krka, d. d. (Novo Mesto, Slovenia). Polyvinylpyrrolidone, grade K-30 (PVP, marketed as Kollidon 30), was sourced from BASF SE (Ludwigshafen, Germany). Microcrystalline cellulose, grade 101 (MCC, marketed as Vivapur® 101), was sourced from JRS PHARMA GmbH & Co. KG (Rosenberg, Germany). Ethanol was supplied by Krka, d. d. (Slovenia).

Granule Preparation

The target granulate formulation consisted of 19.0 w/w% AM, 19.0 w/w% PVP, and 62.0 w/w% MCC. The granulation liquid was prepared by dissolving AM and PVP in 96 v/v% ethanol by heating to 50 °C. AM and PVP together composed 14 w/w% of the granulation liquid.

The batches were prepared with a laboratory-scale FB granulator—Multilab GPCG (Glatt GmbH, Binzen, Germany), equipped with a peristaltic pump and a 1.2 mm nozzle for spraying the granulation liquid. Process conditions were as follows: inlet air temperature of 55–65 °C, inlet airflow of 50–80 m³/h, spray rate of 15–35 g/min, and atomizing air pressure of 1.5 bar. The granulation liquid spray rate and inlet airflow were manually adjusted during the process to maintain a target product temperature of 32–34 °C, ensuring the optimal fluidization of the product. A total of 1000 g of MCC was utilised as the starting material for the granulations.

Svetič, S.; Medved, L.; Korasa, K.; Vrečer, F. Quantification of Amlodipine Maleate Content in Amorphous Solid Dispersions Produced by Fluidized Bed Granulation Using Process Analytical Technology Tools. Pharmaceutics 2024, 16, 1538. https://doi.org/10.3390/pharmaceutics16121538