In-situ monitoring of in vitro drug release processes in tablets using optical coherence tomography.

Film-coated modified-release tablets are an important dosage form amenable to targeted, controlled, or delayed drug release in the specific region of the gastrointestinal (GI) tract. Depending on the film composition and interaction with the GI fluid, such coated products can modulate the local bioavailability, systemic absorption, protection as an enteric barrier, etc. Although the interaction of a dosage form with the surrounding dissolution medium is vital for the resulting release behavior, the underlying physicochemical phenomena at the film and core levels occurring during the drug release process have not yet been well described. In this work, we attempted to tackle this limitation by introducing a novel in vitro test based on optical coherence tomography (OCT) that allows an in-situ investigation of the sub-surface processes occurring during the drug release. Using a commercially available tablet based on osmotic-controlled release oral delivery systems (OROS), we demonstrated the performance of the presented prototype in terms of monitoring the membrane thickness and thickness variability, the surface roughness, the core swelling behavior, and the porosity of the core matrix throughout the in vitro drug release process from OROS. The superior spatial (micron scale) and temporal (less than 10 ms between the subsequent tomograms) resolution achieved in the proposed setup provides an improved understanding of the dynamics inside the microstructure at any given time during the dissolution procedure with the previously unattainable resolution, offering new opportunities for the design and testing of patient-centric dosage forms.

Extending the Use of Optical Coherence Tomography to Scattering Coatings Containing Pigments.

Coating thickness is a critical quality attribute of many coated tablets. Functional coatings ensure correct drug release kinetics or protection from light, while non-functional coatings are generally applied for cosmetic reasons. Traditionally, coating thickness is assessed indirectly via offline methods, such as weight gain or diameter growth. In the past decade, several methods, including optical coherence tomography (OCT) and Raman spectroscopy, have emerged to perform in-line measurements of various subclasses of coating formulations. However, there are some obstacles. For example, when using OCT, a major challenge is scattering pigments, such as titanium dioxide and iron oxide, which make the interface between the coating and the tablet core difficult to detect. This work explores novel OCT image evaluation techniques using unsupervised machine learning to compute image metrics. Certain image metrics of highly scattering coatings are correlated with the tablet thickness, and hence indirectly with the coating thickness. The method was demonstrated using a titanium dioxide rich coating formulation. The results are expected to be applicable to other scattering coatings and will significantly broaden the applicability of OCT to at-line and in-line coating thickness measurements of a much larger class of coating formulations.

Single-tablet-scale direct-compression: An on-demand manufacturing route for personalized tablets.

Today's pharmaceutical industry is facing various challenges. Two of them are issues with supply chain security and the increasing demand for personalized medicine. Both can be addressed by increasing flexibility and a more decentralized approach to pharmaceutical manufacturing. In this study, we present a setup that provides flexibility in terms of supplied raw materials and the product, i.e., a direct-compression setup for personalized tablets operating at a single-tablet-scale. The performance of the implemented single-tablet-scale technology for dosing and mixing was investigated. In addition, an analysis of the critical quality attributes (CQAs) of immediate release ibuprofen and loratadine tablets was performed. The developed dosing device achieved acceptance rates of > 90 % for doses ≥ 20 mg for various pharmaceutical powders. Regarding the vibratory mixing process, a dependency of the performance on the applied frequencies and acceleration was observed, with 100 Hz and âˆ¼ 90 G performing best, yet still exhibiting varying mixing efficacies depending on the granular system. The tablets produced met U.S. Pharmacopeia requirements regarding mechanical stability and dissolution characteristics. Given these results, we consider the developed setup a proof of concept of a tool to provide personalized tablets to patients while minimizing the dependency on complex supply chains.

A multi-step machine learning approach for accelerating QbD-based process development of protein spray drying.

This study proposes a new material-efficient multi-step machine learning (ML) approach for the development of a design space (DS) for spray drying proteins. Typically, a DS is developed by performing a design of experiments (DoE) with the spray dryer and the protein of interest, followed by deriving the DoE models via multi-variate regression. This approach was followed as a benchmark to the ML approach. The more complex the process and required accuracy of the final model is, the more experiments are necessary. However, most biologics are expensive and thus experiments should be kept to a minimum. Therefore, the suitability of using a surrogate material and ML for the development of a DS was investigated. To this end, a DoE was performed with the surrogate and the data used for training the ML approach. The ML and DoE model predictions were compared to measurements of three protein-based validation runs. The suitability of using lactose as surrogate was investigated and advantages of the proposed approach were demonstrated. Limitations were identified at protein concentrations >35 mg/ml and particle sizes of x50>6 µm. Within the investigated DS protein secondary structure was preserved, and most process settings, resulted in yields >75% and residual moisture <10 wt%.

A DEM model to evaluate refill strategies of a twin-screw feeder.

Residence time distribution (RTD) modeling has proven to be a valuable tool for material tracking in continuous pharmaceutical processes. Refilling is thoroughly studied in the literature, but the main focus lies on the feed rate disturbances. The impact of the feeders themselves on intermixing of different material batches is often overlooked. Since the experimental methods to measure the RTD feeder discharging processes feeder are complex and material intensive, there is only limited experimental RTD data available in the literature. A DEM (discrete element method) simulation of a discharge of a twin-screw feeder shows that a large fraction of material that is moved and intermixed by the agitator. In addition to the intermixing, there is a tendency to discharge material located above the agitator early. In order to predict the behavior during multiple refill events, three models in order of increasing complexity are presented: (1) A simple exponential RTD assuming perfect intermixing of material batches; (2) a RTD model based on DEM results; (3) particle-level material tracking by extrapolation of the DEM results. All three of these models are able to predict the survival function of old material for late refills at low fill levels, however, earlier refills at high fill levels require more complex models to accurately represent the dynamics inside the hopper of the feeder.

Predicting powder feedability: A workflow for assessing the risk of flow stagnation and defining the operating space for different powder-feeder combinations.

Powder feeding is of critical importance for continuous manufacturing (CM) since next to in-process segregation it is the phenomenon primarily responsible for fluctuations in content uniformity and for content deviations in the final drug product. So far, feeding studies have focused on the characterization of specific feeders and the prediction of their performance for various materials. This work presents a more holistic approach, an early general assessment of the "feedability" of raw materials. With that regard, we established a workflow to: i) predict potential feeding issues, such as the flow stagnation in the hopper based on both the material attributes and the feeder's geometry; and ii) predict the feed rate space using various feeder/screw combinations for powders with an acceptable risk of hopper flow stagnation. Statistical models were developed for this twofold approach using a dataset comprising nine powders and four different feeders. In order to include different feeding equipment into the statistical models, novel equipment descriptors (capturing the effect of different geometries) and performance indicators (the end fill level as indicator for the risk of powder flow stagnation) were introduced. The application of the workflow was demonstrated for a simple formulation, and model validation was successfully performed for an additional powder that was not contained in the original dataset. Finally, the most relevant material attributes were identified, and reduced material characterization data sets were investigated in terms of effects on the model's prediction performance. The workflow presents a promising tool for initial process assessment in early-phase development.

Ascertain a minimum coating thickness for acid protection of enteric coatings by means of optical coherence tomography.

Enteric coatings are designed to protect active pharmaceutical ingredients (APIs) against untimely release in the stomach. Acid protection of such coatings depends on the coating layer thickness and integrity, which must be determined in an accurate and reliable way to ensure the final product's desired performance. Our work addresses the use of optical coherence tomography (OCT) for characterizing the coating thickness and variability of an enteric-coated drug product and linking them to resistance against gastric fluid. In this study, three batches of enteric-coated tablets drawn during the manufacturing process were investigated. An industrial OCT system was used to establish the coating thickness variability of single tablets (intra-tablet), all tablets in a batch (inter-tablet) and between the batches (inter-batch). Based on the large amount of OCT data, we calculated a critical coating thickness for the investigated film coating, which was found to be 27.4 µm. The corresponding distribution has a mean coating thickness of 44.3 µm ± 7.8 µm. The final coated product has a final mean coating thickness of 63.4 µm ± 8.7 µm, guaranteeing that all tablets meet the quality criterion (i.e., acid protection). Based on the measured thickness distributions, already known distribution functions were considered and an additional, new function was proposed for characterizing the coating thickness distributions in the early stages of industrial coating processes. The proposed approach can be transferred to in-line monitoring of the tablet coating processes, which could drastically improve the production efficiency by ultimately allowing real-time release testing (RTRT).

Carrier particle emission and dispersion in transient CFD-DEM simulations of a capsule-based DPI.

The dispersion of carrier-based formulations in capsule-based dry powder inhalers depends on several factors, including the patient's inhalation profile and the motion of capsule within the device. In the present study, coupled computational fluid dynamics and discrete element method simulations of a polydisperse cohesive lactose carrier in an Aerolizer® inhaler were conducted at a constant flow rate of 100 L/min and considering an inhalation profile of asthmatic children between 5 and 17 years approximated from literature data. In relevant high-speed photography experiments, it was observed that the powder was distributed to both capsule ends before being ejected from the capsule. Several methods of ensuring similar behavior in the simulations were presented. Both the constant flow rate simulation and the profile simulations showed a high powder retention in the capsule (7.37-19.00%). Although the inhaler retention was negligible in the constant flow rate simulation due to consistently high air velocities in the device, it reached values of around 7% in most of the profile simulations. In all simulations, some of the carrier powder was ejected from the capsule as particle clusters. These clusters were larger in the profile simulation than in the constant flow rate simulation. Of the powder discharged from the capsule, a high percentage was bound in clusters in the profile simulation in the beginning and at the end of the inhalation profile while no more than 10% of the powder ejected from the capsule in the 100 L/min constant flow rate simulation were in clusters at any time. The powder emission from the capsule was studied, indicating a strong dependency of the powder mass flow from the capsule on the angular capsule position. When the capsule holes face the inhaler's air inlets, the air flow into the capsule restricts the powder discharge. The presented results provide a detailed view of some aspects of the powder flow and dispersion of a cohesive carrier in a capsule-based inhaler device. Furthermore, the importance of considering inhalation profiles in addition to conventional constant flow rate simulations was confirmed.

PAT implementation for advanced process control in solid dosage manufacturing - A practical guide.

The implementation of continuous pharmaceutical manufacturing requires advanced control strategies rather than traditional end product testing or an operation within a small range of controlled parameters. A high level of automation based on process models and hierarchical control concepts is desired. The relevant tools that have been developed and successfully tested in academic and industrial environments in recent years are now ready for utilization on the commercial scale. To date, the focus in Process Analytical Technology (PAT) has mainly been on achieving process understanding and quality control with the ultimate goal of real-time release testing (RTRT). This work describes the workflow for the development of an in-line monitoring strategy to support PAT-based real-time control actions and its integration into solid dosage manufacturing. All stages are discussed in this paper, from process analysis and definition of the monitoring task to technology assessment and selection, its process integration and the development of data acquisition.

X-ray imaging: A potential enabler of automated particulate detection and cake-structure analysis in lyophilized products?

The presence of particulate matter in parenteral products is a major concern since it affects the patients' safety and is one of the main reasons for product recalls. Conventional quality control is based on a visual inspection, which is a labour-intensive task. Limited to clear solutions and the surface of lyophilised products, it cannot be applied to opaque containers. This study assesses the application of X-ray imaging for detecting the particulate matter in a pharmaceutical lyophilized product. The most common types of particulates (i.e., steel, glass, lyo stopper, polymers and organics in different size classes) were intentionally spiked in vials. After optimizing all relevant parameters of the X-ray set-up, all classes of particulates were detected. At the same time, due to contrast enhancement, the inherent structures of lyophilized cake became obvious. This work addresses the potential and limits of X-ray technology in that regard, paving the way for automated image-based particulate matter detection. Moreover, this paper discusses using this approach to predict critical quality attributes (CQAs) of the drug product based on the cake structure attributes.

Development of a Controlled Continuous Low-Dose Feeding Process.

This paper proposes a feed rate control strategy for a novel volumetric micro-feeder, which can accomplish low-dose feeding of pharmaceutical raw materials with significantly different powder properties. The developed feed-forward control strategy enables a constant feed rate with a minimum deviation from the set-point, even for materials that are typically difficult to accurately feed (e.g., due to high cohesion or low density) using conventional continuous feeders. Density variations observed during the feeding process were characterized via a displacement feed factor profile for each powder. The characterized effective displacement density profile was applied in the micro-feeder system to proactively control the feed rate by manipulating the powder displacement rate (i.e., computing the feed rate from the powder displacement rate). Based on the displacement feed factor profile, the feed rate can be predicted during the feeding process and at any feed rate set-point. Three pharmaceutically relevant materials were used for the micro-feeder evaluation: di-calcium phosphate (large-particle system, high density), croscarmellose sodium (small-particle system, medium density), and barium sulfate (very small-particle <10 µm, high density). A significant improvement in the feeding performance was achieved for all investigated materials. The feed rate deviation from the set-point and its relative standard deviation were minimal compared to operations without the control strategy.

Near-Infrared Hyperspectral Imaging as a Monitoring Tool for On-Demand Manufacturing of Inkjet-Printed Formulations.

This study evaluates the potential use of near-infrared hyperspectral imaging (NIR-HSI) for quantitative determination of the drug amount in inkjet-printed dosage forms. We chose metformin hydrochloride as a model active pharmaceutical ingredient (API) and printed it onto gelatin films using a piezoelectric inkjet printing system. An industry-ready NIR-HSI sensor combined with a motorized movable linear stage was applied for spectral acquisition. Initial API-substrate screening revealed best printing results for gelatin films with TiO2 filling. For calibration of the NIR-HSI system, escalating drug doses were printed on the substrate. After spectral pre-treatments, including standard normal variate (SNV) and Savitzky-Golay filtering for noise reduction and enhancement of spectral features, principal component analysis (PCA) and partial least squares (PLS) regression were applied to create predictive models for the quantification of independent printed metformin hydrochloride samples. It could be shown that the concentration distribution maps provided by the developed HSI models were capable of clustering and predicting the drug dose in the formulations. HSI model prediction showed significant better correlation to the reference (HPLC) compared to on-board monitoring of dispensed volume of the printer. Overall, the results emphasize the capability of NIR-HSI as a fast and non-destructive method for the quantification and quality control of the deposited API in drug-printing applications.

Feasibility of In-line monitoring of critical coating quality attributes via OCT: Thickness, variability, film homogeneity and roughness.

The feasibility of Optical Coherence Tomography (OCT) for in-line monitoring of pharmaceutical film coating processes has recently been demonstrated. OCT enables real-time acquisition of high-resolution cross-sectional images of coating layers and computation of coating thickness. In addition, coating quality attributes can be computed based on in-line data. This study assesses the in-line applicability of OCT to various coating functionalities and formulations. Several types of commercial film-coated tablets containing the most common ingredients were investigated. To that end, the tablets were placed into a miniaturized perforated drum. An in-line OCT system was used to monitor the tablet bed. This set-up resembles the final stage of an industrial pan coating process. All investigated coatings were measured, and the coating thickness, homogeneity and roughness were computed. The rotation rate was varied in a range comparable to large-scale coating operations, and no influence on the outcome was observed. The results indicate that OCT can be used to determine end-point and establish in-process control for a wide range of coating formulations. The real-time computation of coating homogeneity and roughness can support process optimization and formulation development.

Estimating inter-patient variability of dispersion in dry powder inhalers using CFD-DEM simulations.

Drug delivery from a capsule-based dry powder inhaler depends on the inhaler's design, the drug's formulation, and the inhalation maneuver. The latter affects both the air flow and the capsule motion in the inhaler. It is well known that patient-to-patient variability is a major challenge in the design of new inhaler types. Modeling and simulation are important tools for understanding such systems, yet quite complex. Simulation studies of capsule-based dry powder inhalers have disregarded the transient nature of the inhalation process, adopting a constant flow rate through the inhaler instead. In addition, either no capsules, a capsule in a fixed position, or a capsule rotating at a constant rate have been considered. In this work, literature data for three inhalation flow profiles were incorporated into coupled simulations of the air flow and carrier particle motion through an Aerolizer® dry powder inhaler with a rotating capsule and compared to simulations at constant air flow rates. The results for the profile simulations indicated that the carrier powder experienced larger velocity fluctuations. Acceleration events were tracked as a measure of collision- and flow-induced dispersion. The majority of fast particle accelerations occurred when the particles collided with the swirl chamber walls. Of the two common particle dispersion metrics, only the peak particle force distribution appeared to be sensitive to the inhalation profiles, while the effect of the profiles on the cumulative impulse was small.

Deep convolutional neural networks: Outperforming established algorithms in the evaluation of industrial optical coherence tomography (OCT) images of pharmaceutical coatings.

This paper presents a novel evaluation approach for optical coherence tomography (OCT) image analysis of pharmaceutical solid dosage forms based on deep convolutional neural networks (CNNs). As a proof of concept, CNNs were applied to image data from both, in- and at-line OCT implementations, monitoring film-coated tablets as well as single- and multi-layered pellets. CNN results were compared against results from established algorithms based on ellipse-fitting, as well as to human-annotated ground truth data. Performance benchmarks used include, efficiency (computation speed), sensitivity (number of detections from a defined test set) and accuracy (deviation from the reference method). The results were validated by comparing the output of several algorithms to data manually annotated by human experts and microscopy images of cross-sectional cuts of the same dosage forms as a reference method. In order to guarantee comparability for all results, the algorithms were executed on the same hardware. Since modern OCT systems must operate under real-time conditions in order to be implemented in-line into manufacturing lines, the necessary steps are discussed on how to achieve this goal without sacrificing the algorithmic performance and how to tailor a deep CNN to cope with the high amount of image noise and alterations in object appearance. The developed deep learning approach outperforms static algorithms currently available in pharma applications with respect to performance benchmarks, and represents the next level in real time evaluation of challenging industrial OCT image data.

Validating a Numerical Simulation of the ConsiGma(R) Coater.

Continuous manufacturing is increasingly used in the pharmaceutical industry, as it promises to deliver better product quality while simultaneously increasing production flexibility. GEA developed a semi-continuous tablet coater which can be integrated into a continuous tableting line, accelerating the switch from traditional batch production to the continuous mode of operation. The latter offers certain advantages over batch production, e.g., operational flexibility, increased process/product quality, and decreased cost. However, process understanding is the key element for process control. In this regard, computational tools can improve the fundamental understanding and process performance, especially those related to new processes, such as continuous tablet coating where process mechanics remain unclear. The discrete element method (DEM) and computational fluid dynamics (CFD) are two methods that allow transition from empirical process design to a mechanistic understanding of the individual process units. The developed coupling model allows to track the heat, mass, and momentum exchange between the tablet and fluid phase. The goal of this work was to develop and validate a high-fidelity CFD-DEM simulation model of the tablet coating process in the GEA ConsiGma® coater. After the model development, simulation results for the tablet movement, coating quality, and heat and mass transfer during the coating process were validated and compared to the experimental outcomes. The experimental and simulation results agreed well on all accounts measured, indicating that the model can be used in further studies to investigate the operating space of the continuous tablet coating process.

Performance Evaluation of a High-Precision Low-Dose Powder Feeder.

Highly potent active pharmaceutical ingredients (APIs) and low-dose excipients, or excipients with very low density, are notoriously hard to feed with currently available commercial technology. The micro-feeder system presented in this work is capable of feeding low-dose rates of powders with different particle sizes and flow properties. Two different grades of lactose, di-calcium phosphate, croscarmellose sodium, silicon dioxide, a spray-dried intermediate, and an active ingredient were studied to vary material properties to test performance of the system. The current micro-feeder system is a volumetric feeder combined with a weighing balance at the outlet that measures feeder output rates. Feeding results are shown as a so-called "displacement-feed factor" curve for each material. Since the powder mass and volume are known in the micro-feeder system, in this work, we characterized an observed density variation during processing via a "displacement-feed factor" profile for each of the fed powders. This curve can be later used for calibrating the system to ensure an accurate, constant feed rate and in addition predicting feeding performance for that material at any feed rate. There is a relation between powder properties and feeding performance. Powders with finer particles and higher compressibility show densification during their feeding process. However, powders with larger particles and lower compressibility show both "densification" and "powder bed expansion," which is the manifestation of dilation and elastic recovery of particles during the micro-feeding process. Through the application of the displacement-feed factor, it is possible to provide precise feeding accuracy of low-dose materials.

A solution for low-dose feeding in continuous pharmaceutical processes.

Continuous feeding of small quantities of powder is increasingly applied in pharmaceutical manufacturing. With that regard, what is crucial is not only the feasibility, but also the accuracy and stability. To enable stable processing, low amounts of various agents, e.g., lubricants, can be used. Even more important is the exact dosage of highly potent active pharmaceutical ingredients (APIs), which require feed rates within the range of grams per hour. Conventional feeders cannot supply powder at such rates, especially when the material properties are challenging. In this work, a novel micro-feeder was integrated into a continuous manufacturing line and its capability to supply API at feed rates down to one gram per hour was tested. The micro-feeder system is based on the principle of active volumetric displacement: a piston pushes the powder out of the cartridge upwards to the end of a plate, where a scraper places it into the process inlet. In this study, a hot melt extrusion process was used, during which the API was dissolved in a polymer matrix. Samples of the strand were analysed with regard to their content by means of HPLC. The results showed that the novel micro-feeder system can feed powder with good accuracy and reproducibility, indicating its high potential for continuous process implementation.

Developing HME-Based Drug Products Using Emerging Science: a Fast-Track Roadmap from Concept to Clinical Batch.

This paper presents a rational workflow for developing enabling formulations, such as amorphous solid dispersions, via hot-melt extrusion in less than a year. First, our approach to an integrated product and process development framework is described, including state-of-the-art theoretical concepts, modeling, and experimental characterization described in the literature and developed by us. Next, lab-scale extruder setups are designed (processing conditions and screw design) based on a rational, model-based framework that takes into account the thermal load required, the mixing capabilities, and the thermo-mechanical degradation. The predicted optimal process setup can be validated quickly in the pilot plant. Lastly, a transfer of the process to any GMP-certified manufacturing site can be performed in silico for any extruder based on our validated computational framework. In summary, the proposed workflow massively reduces the risk in product and process development and shortens the drug-to-market time for enabling formulations.

Filling of lactose-based formulations in a tamping-pin capsule filler.

We studied three lactose-based formulations in terms of bulk powder properties and capsule-filling behavior in a tamping-pin capsule filling system, to which several mechanical adaptions were made for process optimization in light of future continuous production. The model formulations were thoroughly characterized and filled into size 1 capsules according a well-defined design of experiments (DoE). Overall, the three entirely different formulations were successfully filled within the selected design space. The fill weight and fill weight variability can be adjusted by fine-tuning the process settings, like the pin immersion depth and the maximum compaction pressure (pneumatic or spring-controlled), and by using the appropriate powder bed height and mechanical adaptions. This study demonstrated that selection of process parameters and mechanical adaptions could enhance the filling performance, especially in continuous production, since they reduce the powder volume in the process. Moreover, we showed that a tamping-pin system is capable of successfully filling a broad range of powders with various material characteristics and can potentially be used in a continuous production mode.