PAT-based batch statistical process control of a manufacturing process for a pharmaceutical ointment.

In this study, a process analytical technology (PAT)-based batch statistical process control (BSPC) model was developed for the laboratory-scale manufacturing process of a commercially available pharmaceutical ointment. The multivariate BSPC model was developed based on the in-line measured viscosity (viscometer), product temperature (viscometer), particle size distribution (PSD) (focused beam reflectance measurement (FBRM)) and active pharmaceutical ingredient (API) concentration (Raman spectroscopy) of four reference batches using a partial least squares (PLS) approach. From this in-line collected data, the characteristic trajectory of the batch process under normal operating conditions was acquired. To assess the capability of the process analyzers and BSPC model to detect deviations from the expected batch trajectory, two test batches with induced process and formulation disturbances were monitored in-line. The elevated process temperature in test batch 1 resulted in a deviating viscosity, product temperature and number of small particles (<100 µm). After correcting the process temperature, the viscosity and product temperature were within the control interval, while the particle size was smaller compared to the reference batches. For test batch 2, API was added at three different time points, whereas the same amount of API was added in one step during manufacturing of the reference batches. The induced disturbance was reflected in the in-line measured viscosity, PSD and API concentration. The combination of process analyzers and multivariate batch modelling enabled early fault detection and real-time process adjustments, thereby preventing batch loss or reprocessing. In addition, the feasibility of the investigated process analyzers to measure certain quality attributes in-line during manufacturing of an ointment was demonstrated.

Continuous manufacturing of a pharmaceutical cream: Investigating continuous powder dispersing and residence time distribution (RTD).

Recently, an innovative continuous manufacturing technology for a pharmaceutical oral suspension was proposed, based on two consecutive mixing units. A limitation of this technology is the need to dissolve or disperse powder-based raw materials in a liquid via a batch step before continuous manufacturing. Therefore, the aim of the current study was to develop and investigate a method to introduce powders continuously into the existing equipment via the implementation of two upstream continuous unit operations: a powder feeder and powder dispersing unit. A pharmaceutical cream was selected as model formulation to demonstrate the flexibility of the continuous manufacturing technology towards different types of semi-solid and liquid formulations. The ability to continuously feed and disperse active pharmaceutical ingredient (API) using the proposed method was assessed via an experimental design, in which the impact of several process parameters of the powder dispersing unit on the API concentration (relative error (RE) and relative standard deviation (RSD)) was examined. A Raman spectroscopic method was developed to quantify the API concentration in-line after the powder dispersing step. The API concentration was independent of the process parameters and fell within the acceptance limits, except for two experimental runs where a deviating API concentration was observed. These results demonstrate that the continuous powder feeding and dispersing method was suitable, and that a completely continuous manufacturing system was obtained. To achieve raw material traceability and understanding the mixing behavior, the residence time distribution (RTD) of a tracer inside the continuous manufacturing equipment was determined using a colorimetric technique. The time required to remove all tracer from the powder dispersing unit operation was very long (1481 s) and therefore the volume inside this unit operation should be reduced by designing new equipment with smaller dimensions. At the two consecutive mixing units, the peak and mean residence time were influenced by throughput, whereas mixing speed in both mixing units had a significant impact on the degree of axial mixing. Finally, the continuously manufactured cream had a similar rheological behavior as the original batch-wise manufactured cream.

A multivariate approach to predict the volumetric and gravimetric feeding behavior of a low feed rate feeder based on raw material properties.

In this study, the volumetric and gravimetric feeding behavior of 15 pharmaceutical powders on a low feed rate feeder was correlated with their material properties through a multivariate approach. The powders under investigation differ substantially in terms of material properties, making the selected powders representative for powders typically used in pharmaceutical manufacturing. The material properties were described by 25 material property descriptors, obtained from a rational selection of critical characterization techniques that provided maximal information with minimal characterization effort. From volumetric feeding experiments (i.e., powder feed rate not controlled), the maximum feeding capacity (maximum feed factor (FFmax)) and optimal hopper fill level at which the feeder should be refilled during gravimetric feeding (feed factor decay (FFdecay)) were obtained. During gravimetric feeding experiments (i.e., powder feed rate controlled), the variability on the feed rate (relative standard deviation (RSD)) and the difference between the setpoint and mean feed rate (relative error (RE)) were determined. Partial least squares (PLS) regression was applied to correlate the volumetric and gravimetric feeding responses (Y) with the material property descriptors (X). The predictive ability of the developed PLS models was assessed by predicting the feeding responses of two new powders (i.e., validation set). Overall, the volumetric feeding responses (FFmax and FFdecay) were predicted better than the gravimetric feeding responses (RSD and RE), since in gravimetric mode the impact of material properties on the feeding behavior is reduced due to the control system of the feeder. Especially RE was weakly correlated with material properties as RE of most powders varied around zero with only a small numerical variation. Interestingly, this confirms that the control system is working properly and that the feeder is capable of feeding different powders accurately at low feed rates. The developed models allowed to predict the feeding behavior of new powders based on their material properties. Consequently the number of feeding experiments during process development can be greatly reduced, thereby leading to a more efficient and faster development of new drug products.

A continuous manufacturing concept for a pharmaceutical oral suspension.

The aim of this study was to investigate the applicability of an innovative continuous manufacturing system for semi-solid and liquid pharmaceutical formulations. A commercially available pharmaceutical oral suspension was selected as model formulation. Premixes of the raw materials were dosed via peristaltic pumps to the mixing compartment, which consists of two consecutive mixing units. An experimental design was used to study the influence of several process parameters (throughput, mixing speed in mixing unit 1 and mixing speed in mixing unit 2) on the quality attributes of the oral suspension. The pH, density, active pharmaceutical ingredient (API) concentration, sedimentation after 30 days (expressed by the sedimentation volume) and rheological characteristic (yield stress) of the suspension were determined. No significant influence of the process parameters on the pH, density and API concentration was observed. The throughput and mixing speed in mixing unit 1 had a significant impact on both the sedimentation volume and yield stress, and were therefore critical to acquire physical stable suspensions. Furthermore, the yield stress measured one day after production was predictive for the occurrence of sedimentation in the suspensions after 30 days. When selecting the optimal process settings, the continuously manufactured suspension had a similar product quality as the original batch-processed suspension and even possessed a higher yield stress. This study demonstrated that the investigated innovative continuous manufacturing technology is suitable for the manufacturing of a commercially available pharmaceutical suspension and that the product quality can be optimized by adjusting the process parameters.

In-line UV spectroscopy for the quantification of low-dose active ingredients during the manufacturing of pharmaceutical semi-solid and liquid formulations.

UltraViolet (UV) spectroscopy was evaluated as an innovative Process Analytical Technology (PAT) - tool for the in-line and real-time quantitative determination of low-dosed active pharmaceutical ingredients (APIs) in a semi-solid (gel) and a liquid (suspension) pharmaceutical formulation during their batch production process. The performance of this new PAT-tool (i.e., UV spectroscopy) was compared with an already more established PAT-method based on Raman spectroscopy. In-line UV measurements were carried out with an immersion probe while for the Raman measurements a non-contact PhAT probe was used. For both studied formulations, an in-line API quantification model was developed and validated per spectroscopic technique. The known API concentrations (Y) were correlated with the corresponding in-line collected preprocessed spectra (X) through a Partial Least Squares (PLS) regression. Each developed quantification method was validated by calculating the accuracy profile on the basis of the validation experiments. Furthermore, the measurement uncertainty was determined based on the data generated for the determination of the accuracy profiles. From the accuracy profile of the UV- and Raman-based quantification method for the gel, it was concluded that at the target API concentration of 2% (w/w), 95 out of 100 future routine measurements given by the Raman method will not deviate more than 10% (relative error) from the true API concentration, whereas for the UV method the acceptance limits of 10% were exceeded. For the liquid formulation, the Raman method was not able to quantify the API in the low-dosed suspension (0.09% (w/w) API). In contrast, the in-line UV method was able to adequately quantify the API in the suspension. This study demonstrated that UV spectroscopy can be adopted as a novel in-line PAT-technique for low-dose quantification purposes in pharmaceutical processes. Important is that none of the two spectroscopic techniques was superior to the other for both formulations: the Raman method was more accurate in quantifying the API in the gel (2% (w/w) API), while the UV method performed better for API quantification in the suspension (0.09% (w/w) API).

Breakage and drying behaviour of granules in a continuous fluid bed dryer: Influence of process parameters and wet granule transfer.

Although twin screw granulation has already been widely studied in recent years, only few studies addressed the subsequent continuous drying which is required after wet granulation and still suffers from a lack of detailed understanding. The latter is important for optimisation and control and, hence, a cost-effective practical implementation. Therefore, the aim of the current study is to increase understanding of the drying kinetics and the breakage and attrition phenomena during fluid bed drying after continuous twin screw granulation. Experiments were performed on a continuous manufacturing line consisting of a twin-screw granulator, a six-segmented fluid bed dryer, a mill, a lubricant blender and a tablet press. Granulation parameters were fixed in order to only examine the effect of drying parameters (filling time, drying time, air flow, drying air temperature) on the size distribution and moisture content of granules (both of the entire granulate and of size fractions). The wet granules were transferred either gravimetrically or pneumatically from the granulator exit to the fluid bed dryer. After a certain drying time, the moisture content reached an equilibrium. This drying time was found to depend on the applied airflow, drying air temperature and filling time. The moisture content of the granules decreased with an increasing drying time, airflow and drying temperature. Although smaller granules dried faster, the multimodal particle size distribution of the granules did not compromise uniform drying of the granules when the target moisture content was achieved. Extensive breakage of granules was observed during drying. Especially wet granules were prone to breakage and attrition during pneumatic transport, either in the wet transfer line or in the dry transfer line. Breakage and attrition of granules during transport and drying should be anticipated early on during process and formulation development by performing integrated experiments on the granulator, dryer and mill.

Downstream processing from hot-melt extrusion towards tablets: A quality by design approach.

Since the concept of continuous processing is gaining momentum in pharmaceutical manufacturing, a thorough understanding on how process and formulation parameters can impact the critical quality attributes (CQA) of the end product is more than ever required. This study was designed to screen the influence of process parameters and drug load during HME on both extrudate properties and tableting behaviour of an amorphous solid dispersion formulation using a quality-by-design (QbD) approach. A full factorial experimental design with 19 experiments was used to evaluate the effect of several process variables (barrel temperature: 160-200°C, screw speed: 50-200rpm, throughput: 0.2-0.5kg/h) and drug load (0-20%) as formulation parameter on the hot-melt extrusion (HME) process, extrudate and tablet quality of Soluplus®-Celecoxib amorphous solid dispersions. A prominent impact of the formulation parameter on the CQA of the extrudates (i.e. solid state properties, moisture content, particle size distribution) and tablets (i.e. tabletability, compactibility, fragmentary behaviour, elastic recovery) was discovered. The resistance of the polymer matrix to thermo-mechanical stress during HME was confirmed throughout the experimental design space. In addition, the suitability of Raman spectroscopy as verification method for the active pharmaceutical ingredient (API) concentration in solid dispersions was evaluated. Incorporation of the Raman spectroscopy data in a PLS model enabled API quantification in the extrudate powders with none of the DOE-experiments resulting in extrudates with a CEL content deviating>3% of the label claim. This research paper emphasized that HME is a robust process throughout the experimental design space for obtaining amorphous glassy solutions and for tabletting of such formulations since only minimal impact of the process parameters was detected on the extrudate and tablet properties. However, the quality of extrudates and tablets can be optimized by adjusting specific formulations parameters (e.g. drug load).

Multivariate statistical process control of a continuous pharmaceutical twin-screw granulation and fluid bed drying process.

A multivariate statistical process control (MSPC) strategy was developed for the monitoring of the ConsiGma™-25 continuous tablet manufacturing line. Thirty-five logged variables encompassing three major units, being a twin screw high shear granulator, a fluid bed dryer and a product control unit, were used to monitor the process. The MSPC strategy was based on principal component analysis of data acquired under normal operating conditions using a series of four process runs. Runs with imposed disturbances in the dryer air flow and temperature, in the granulator barrel temperature, speed and liquid mass flow and in the powder dosing unit mass flow were utilized to evaluate the model's monitoring performance. The impact of the imposed deviations to the process continuity was also evaluated using Hotelling's T2 and Q residuals statistics control charts. The influence of the individual process variables was assessed by analyzing contribution plots at specific time points. Results show that the imposed disturbances were all detected in both control charts. Overall, the MSPC strategy was successfully developed and applied. Additionally, deviations not associated with the imposed changes were detected, mainly in the granulator barrel temperature control.