Hemlata Patil, Senior Manager, Hot Melt Extrusion Technology, Catalent
There are two common techniques for dry granulation used in the pharmaceutical industry: roller compaction and slugging. However, both these methods have some limitations. First, tablets with dry granules prepared via slugging or roller compaction method exhibit an inferior tensile strength and second, product yield can be reduced because of the high amount of fine particulates that the processes create.
To overcome these limitations, there is another method to prepare dry granules: a twin-screw extruder, where the drug-polymer blend is put into the extruder’s feeder, which consists of twin screws with convening elements. Material moves forward into the barrel as a function of conveying screw elements, compresses the blend to form agglomerates with the help of kneading screw elements, and then, at the end of the barrel, the combing screw element helps break down these big agglomerates into small granules. There is no die at the end of the barrel, and the resulting granules are mixed with extra granular material to prepare the final blend, which is then compressed into tablets or filled into capsules or stick packs (Figure 1).
The barrel temperature remains lower than the glass transition temperature or melting point of the excipients. During the process, the mechanism of dry granulation applies pressure to the powder to induce interparticular bonding through various interactions such as Van Der Waals forces and mechanical interlocking. When the surfaces of two particles approach each other closely enough, the free surface energies result in a strong, attractive force through a process known as cold welding. This increases the mechanical strength of the powder bed when subjected to rising compressive forces within the twin screw extruder. If the API is crystalline during this process, there is no change in its form, so the granulated API remains in that state.
Case Study: Using Twin Screw Granulation to Develop a Controlled Release Formulation
The study’s aim was to create a controlled-release version of Ondansetron tablet using a Quality-by-Design (QbD) approach. Ondansetron is a medication given to cancer patients to prevent nausea and vomiting caused by chemotherapy, radiation therapy or surgery. It can also be administered to treat gastroenteritis. The process involved the utilization of Design of Experiment (DoE) as a proactive measure to minimize risks using a twin-screw dry granulation process. The DoE was used to examine how screw speed, feed rate, and fumaric acid concentration affect the drug release profile of the granules. The results of the DoE were used to further develop the process design by varying formulation composition, barrel temperature and screw configuration and optimize the conditions for producing granules with the desired Critical Quality Attributes (CQAs). Ondansetron as an API is a weak base and exhibits pH-dependent solubility, so fumaric acid was also used in the formulation, which acts as an acidifier.
Figure 2 shows the correlation between critical process parameters and critical quality attributes to define design space.
The interaction between formulation variables and extrusion process parameters are complex and require an in-depth knowledge of the extruder and extrusion process. The initial process parameters were divided into independent variables and dependent variables, as shown in Figure 3.
A Quality Target Product Profile (QTPP) had been defined, and the desired dosage form was an orally delivered controlled-release tablet, containing 30mg of API, with at least six months stability at accelerated conditions (40°C / 75% relative humidity). All work was carried out on an 11mm twin screw extruder.
CQAs were defined based on particle size (1.4mm-500μm), the angle of repose (25-31°) and a dissolution time to provide 100% drug release over 12-16 hours. The angle of repose of a granular material is the steepest angle of descent or dip relative to the horizontal plane on which the material can be piled without slumping.
Risk Assessment
The initial risk assessment for the process showed that the screw configuration and screw speed, as well as the flow rate, had a high impact on the resulting granule particle size. The specification of the input material needed to be controlled, as those differences in the particle size of API and excipients could lead to segregation issues in the blend and exaggerated over time without optimal screw speed and feed rate. The risk of unwanted related substances was minimal because the temperature of the process was low compared to other processes, such as milling or compression techniques.
Content uniformity of product was seen as a medium risk for the same reasons as variation in particle size: screw configurations and screw speed flow rate can change the property of the granules. Very porous product granules may enhance the drug release, potentially causing immediate release of API, whereas very dense and low porosity granules may delay release for a long period of time, neither of which were appropriate for the 12-16h dissolution CQA.
So, considering these initial risks, three independent variables were chosen for the DoE design: screw speed (25-100 RPM), feed rate (1-3%) and fumaric acid content (2.5-7.5%), with responses to be measured being particle size, dissolution, and flow properties.
Process Design
Three different formulations were used, each with 20% Ondansetron and fumaric acid at 2.5, 5, and 7.5%. Ethocel® (EC) was added to each as a hydrophobic polymer for the purpose of sustained release, along with hydroxypropylcellulose (HPC) EF as a dry binder in a 1:1 ratio.
Three different screw configurations were evaluated to be used for the experiments, one that was all conveying, as well as configurations featuring single and dual mixing zones. Temperatures remained consistent with each configuration, but varied across the feed zones between 70 and 90°C.
The effect of the screw configuration can be seen in Figure 4. Absence of a kneading zone in the screw configuration resulted in the generation of 20% granules and more than 80% material remained as fine blend even under higher temperatures. When multiple kneading zones were used, large granules formed, but in the configuration with only one kneading zone, these became medium-sized. It was also observed that the granules from the dual kneading zone screw configuration had a rubbery texture and were hard to break. As such, a single kneading zone screw configuration was selected for the DoE studies.
Two3 factorial design was used to conduct ten experiments with various screw speeds, feed rates and formulations, as shown in Table 1.
After execution, the torque values on the extruder were recorded, and utilizing these torque values and screw speed, the specific mechanical energy for each batch could be calculated.
The granule particle size distribution results for all the formulations are shown in Table 2. Three granule size fractions were collected: large granules (>1.4 mm), medium-sized granules (fraction between 1.4 mm and 500 μm), and nongranulated fi ne particles (< 500 μm). For tableting purposes, medium-sized granules containing a fraction between 1.4 mm and 500 μm are of critical importance and the requirement of 10-15% fi ne particles to obtain good compressibility. Therefore, particle sizes between 1.4 mm and 500 μm were selected as the dependent variable (response) in the 23-factorial design. Along with granule size, granule flow properties are known to impact the content uniformity and tablet mechanical characteristics critically as well. Granule flowability was assessed by measuring Hausner’s ratio, Carr’s index, and angle of repose for all the formulations, and the results are summarized in Table 2.
Of the 10 experiments, batch Z8 showed the optimal characteristics of the granules it produced. Only 4.8% of the granules were less than 500 microns, and less than 6% were oversized. The angle of repose sat within the CQA limit at 25°, demonstrating excellent flow characteristics. Additionally, the dis- solution data suggests sustained release could be achieved over 15.5 hours.
Analysis of Results Against CQAs
It was found that all the independent factors - feed rate, screw speed and amount of fumaric acid in the formulation - had significant effects on the particle size of granules, with the most significant factor among these being feed rate. The highest percentage of medium-sized granules were obtained at low feed rate that is 1%, which may be attributed by the amount of material present inside the barrel that then affects the agglomeration step. With high feed rates, the screw is forcing higher amounts of material through the barrel, resulting in denser, and larger particles.
The second CQA was dissolution, and results showed that screw speed proved to be the dominant factor here. As shown in Figure 5, increasing the screw speed led to sustained release, and it could be surmised that this led to the high shear generation, resulting in more binding between drug and polymers. Because of this strong binding, the drug release from the granules slowed down.
The final CQA was the flow characteristics of the product, and again, screw speed showed the most significant effect on the angle of repose of the granules. Increasing screw speed increases the angle of repose of the product, giving granules with improved flowability.
Conclusion
The work showed how a QbD approach could be successfully applied to prepare a sustained release formulation of Ondansetron using a twin-screw dry granulation process, whereby it is possible to control and optimize crucial factors such as particle size easily.
The Future of Twin Screw Extrusion in Pharmaceutical Manufacturing
Twin screw extrusion is a viable method to develop several different profi les of pharmaceutical products and offers several advantages over traditional granulation technologies. It is unique in its simplicity, compared to the currently available methods for dry granulation, and using enhanced process knowledge and QbD approaches, it offers a less complex solution for continuous manufacturing.