Vibrational spectroscopy and microspectroscopy analyzing qualitatively and quantitatively pharmaceutical hot melt extrudates

COSMO Models for the Pharmaceutical Development of Parenteral Drug Formulations

The aqueous solubility of active pharmaceutical ingredients is one of the most important features to be considered during the development of parenteral formulations in the pharmaceutical industry. Computational modelling has become in the last years an integral part of pharmaceutical development. In this context, ab initio computational models, such as COnductor-like Screening MOdel (COSMO), have been proposed as promising tools for the prediction of results without the effective use of resources. Nevertheless, despite the clear evaluation of computational resources, some authors had not achieved satisfying results and new calculations and algorithms have been proposed over the years to improve the outcomes. In the development and production of aqueous parenteral formulations, the solubility of Active Pharmaceutical Ingredients (APIs) in an aqueous and biocompatible vehicle is a decisive step. This work aims to study the hypothesis that COSMO models could be useful in the development of new parenteral formulations, mainly aqueous ones.

Machine Learning for Process Monitoring and Control of Hot-Melt Extrusion: Current State of the Art and Future Directions

In the last few decades, hot-melt extrusion (HME) has emerged as a rapidly growing technology in the pharmaceutical industry, due to its various advantages over other fabrication routes for drug delivery systems. After the introduction of the ‘quality by design’ (QbD) approach by the Food and Drug Administration (FDA), many research studies have focused on implementing process analytical technology (PAT), including near-infrared (NIR), Raman, and UV–Vis, coupled with various machine learning algorithms, to monitor and control the HME process in real time. This review gives a comprehensive overview of the application of machine learning algorithms for HME processes, with a focus on pharmaceutical HME applications. The main current challenges in the application of machine learning algorithms for pharmaceutical processes are discussed, with potential future directions for the industry.

Hot-Melt Extrusion: a Roadmap for Product Development

Hot-melt extrusion has found extensive application as a feasible pharmaceutical technological option over recent years. HME applications include solubility enhancement, taste masking, and sustained drug release. As bioavailability enhancement is a hot topic of today's science, one of the main applications of HME is centered on amorphous solid dispersions. This review describes the most significant aspects of HME technology and its use to prepare solid dispersions as a drug formulation strategy to enhance the solubility of poorly soluble drugs. It also addresses molecular and thermodynamic features critical for the physicochemical properties of these systems, mainly in what concerns miscibility and physical stability. Moreover, the importance of applying the Quality by Design philosophy in drug development is also discussed, as well as process analytical technologies in pharmaceutical HME monitoring, under the current standards of product development and regulatory guidance. Graphical Abstract.

Inline UV/Vis spectroscopy as PAT tool for hot-melt extrusion

Hot-melt extrusion on co-rotating twin screw extruders is a focused technology for the production of pharmaceuticals in the context of Quality by Design. Since it is a continuous process, the potential for minimizing product quality fluctuation is enhanced. A typical application of hot-melt extrusion is the production of solid dispersions, where an active pharmaceutical ingredient (API) is distributed within a polymer matrix carrier. For this dosage form, the product quality is related amongst others to the drug content. This can be monitored on- or inline as critical quality attribute by a process analytical technology (PAT) in order to meet the specific requirements of Quality by Design. In this study, an inline UV/Vis spectrometer from ColVisTec was implemented in an early development twin screw extruder and the performance tested in accordance to the ICH Q2 guideline. Therefore, two API (carbamazepine and theophylline) and one polymer matrix (copovidone) were considered with the main focus on the quantification of the drug load. The obtained results revealed the suitability of the implemented PAT tool to quantify the drug load in a typical range for pharmaceutical applications. The effort for data evaluation was minimal due to univariate data analysis, and in combination with a measurement frequency of 1 Hz, the system is sufficient for real-time data acquisition.

The development of an inline Raman spectroscopic analysis method as a quality control tool for hot melt extruded ramipril fixed-dose combination products

Currently in the pharmaceutical industry, continuous manufacturing is an area of significant interest. In particular, hot-melt extrusion (HME) offers many advantages and has been shown to significantly reduce the number of processing steps relative to a conventional product manufacturing line. To control product quality during HME without process interruption, integration of inline analytical technology is critical. Vibrational spectroscopy (Raman, NIR and FT-IR) is often employed and used for real-time measurements because of the non-destructive and rapid nature of these analytical techniques. However, the establishment of reliable Process Analytical Technology (PAT) tools for HME of thermolabile drugs is challenging. Indeed, the Raman effect is inherently weak and might be subject to interference. Moreover, during HME, heating and photodecomposition can occur and disrupt spectra acquisition. The aim of this research article was to explore the use of inline Raman spectroscopy to characterise a thermolabile drug, ramipril (RMP), during continuous HME processing. Offline measurements by HPLC, LC-MS and Raman spectroscopy were used to characterise RMP and its main degradation product, ramipril-diketopiperazine (RMP-DKP, impurity K). A set of HME experiments together with inline Raman spectroscopic analyses were performed. The feasibility of implementing inline Raman spectroscopic analysis to quantify the level of RMP and RMP-DKP in the extrudate was addressed. Two regions in the Raman spectrum were selected to differentiate RMP and RMP-DKP. When regions were combined, a principle component analysis (PCA) model defined by these two main components (PC 1 = 50.1% and PC 2 = 45%) was established. Using HPLC analyses, we were able to confirm that the PC 1 score was attributed to the level of RMP-DKP, and the PC 2 score was related to the RMP drug content. Investigation of the PCA scatterplot indicated that HME processing temperature was not the only factor causing RMP degradation. Additionally, the plasticiser content, feeding speed and screw rotating speed contributed to RMP degradation during HME processing.

Development of an analytical method for crystalline content determination in amorphous solid dispersions produced by hot-melt extrusion using transmission Raman spectroscopy: A feasibility study

The development of a quantitative method determining the crystalline percentage in an amorphous solid dispersion is of great interest in the pharmaceutical field. Indeed, the crystalline Active Pharmaceutical Ingredient transformation into its amorphous state is increasingly used as it enhances the solubility and bioavailability of Biopharmaceutical Classification System class II drugs. One way to produce amorphous solid dispersions is the Hot-Melt Extrusion (HME) process. This study reported the development and the comparison of the analytical performances of two techniques, based on backscattering and transmission Raman spectroscopy, determining the crystalline remaining content in amorphous solid dispersions produced by HME. Principal Component Analysis (PCA) and Partial Least Squares (PLS) regression were performed on preprocessed data and tended towards the same conclusions: for the backscattering Raman results, the use of the DuoScan™ mode improved the PCA and PLS results, due to a larger analyzed sampling volume. For the transmission Raman results, the determination of low crystalline percentages was possible and the best regression model was obtained using this technique. Indeed, the latter acquired spectra through the whole sample volume, in contrast with the previous surface analyses performed using the backscattering mode. This study consequently highlighted the importance of the analyzed sampling volume.

Global approach for the validation of an in-line Raman spectroscopic method to determine the API content in real-time during a hot-melt extrusion process

Since the Food and Drug Administration (FDA) published a guidance based on the Process Analytical Technology (PAT) approach, real-time analyses during manufacturing processes are in real expansion. In this study, in-line Raman spectroscopic analyses were performed during a Hot-Melt Extrusion (HME) process to determine the Active Pharmaceutical Ingredient (API) content in real-time. The method was validated based on a univariate and a multivariate approach and the analytical performances of the obtained models were compared. Moreover, on one hand, in-line data were correlated with the real API concentration present in the sample quantified by a previously validated off-line confocal Raman microspectroscopic method. On the other hand, in-line data were also treated in function of the concentration based on the weighing of the components in the prepared mixture. The importance of developing quantitative methods based on the use of a reference method was thus highlighted. The method was validated according to the total error approach fixing the acceptance limits at ±15% and the α risk at ±5%. This method reaches the requirements of the European Pharmacopeia norms for the uniformity of content of single-dose preparations. The validation proves that future results will be in the acceptance limits with a previously defined probability. Finally, the in-line validated method was compared with the off-line one to demonstrate its ability to be used in routine analyses.

Process analytical techniques for hot-melt extrusion and their application to amorphous solid dispersions

Newly developed active pharmaceutical ingredients (APIs) are often poorly soluble in water. As a result the bioavailability of the API in the human body is reduced. One approach to overcome this restriction is the formulation of amorphous solid dispersions (ASDs), e.g., by hot-melt extrusion (HME). Thus, the poorly soluble crystalline form of the API is transferred into a more soluble amorphous form. To reach this aim in HME, the APIs are embedded in a polymer matrix. The resulting amorphous solid dispersions may contain small amounts of residual crystallinity and have the tendency to recrystallize. For the controlled release of the API in the final drug product the amount of crystallinity has to be known. This review assesses the available analytical methods that have been recently used for the characterization of ASDs and the quantification of crystalline API content. Well-established techniques like near- and mid-infrared spectroscopy (NIR and MIR, respectively), Raman spectroscopy, and emerging ones like UV/VIS, terahertz, and ultrasonic spectroscopy are considered in detail. Furthermore, their advantages and limitations are discussed with regard to general practical applicability as process analytical technology (PAT) tools in industrial manufacturing. The review focuses on spectroscopic methods which have been proven as most suitable for in-line and on-line process analytics. Further aspects are spectroscopic techniques that have been or could be integrated into an extruder.

A simple calibration approach based on film-casting for confocal Raman microscopy to support the development of a hot-melt extrusion process

When developing a new formulation, the development, calibration and validation steps of analytical methods based on vibrational spectroscopy are time-consuming. For each new formulation, real samples must be produced and a "reference method" must be used in order to determine the Active Pharmaceutical Ingredient (API) content of each sample. To circumvent this issue, the paper presents a simple approach based on the film-casting technique used as a calibration tool in the framework of hot-melt extrusion process. Confocal Raman microscopic method was successfully validated for the determination of itraconazole content in film-casting samples. Then, hot-melt extrusion was carried out to produce real samples in order to confront the results obtained with confocal Raman microscopy and Ultra High Performance Liquid Chromatography (UHPLC). The agreement between both methods was demonstrated using a comparison study based on the Bland and Altman's plot.

An investigation into the influence of drug-polymer interactions on the miscibility, processability and structure of polyvinylpyrrolidone-based hot melt extrusion formulations

While hot melt extrusion is now established within the pharmaceutical industry, the prediction of miscibility, processability and structural stability remains a pertinent issue, including the issue of whether molecular interaction is necessary for suitable performance. Here we integrate the use of theoretical and experimental drug-polymer interaction assessment with determination of processability and structure of dispersions in two polyvinylpyrrolidone-based polymers (PVP and PVP vinyl acetate, PVPVA). Caffeine and paracetamol were chosen as model drugs on the basis of their differing hydrogen bonding potential with PVP. Solubility parameter and interaction parameter calculations predicted a greater miscibility for paracetamol, while ATR-FTIR confirmed the hydrogen bonding propensity of the paracetamol with both polymers, with little interaction detected for caffeine. PVP was found to exhibit greater interaction and miscibility with paracetamol than did PVPVA. It was noted that lower processing temperatures (circa 40°C below the Tg of the polymer alone and Tm of the crystalline drug) and higher drug loadings with associated molecular dispersion up to 50% w/w were possible for the paracetamol dispersions, although molecular dispersion with the non-interactive caffeine was noted at loadings up to 20% w./w. A lower processing temperature was also noted for caffeine-loaded systems despite the absence of detectable interactions. The study has therefore indicated that theoretical and experimental detection of miscibility and drug-polymer interactions may lead to insights into product processing and extrudate structure, with direct molecular interaction representing a helpful but not essential aspect of drug-polymer combination prediction.