Manel Bautista, Reto Maurer, Laura Rolinger, Emmanuela Gavi, and Patrick M. Piccione-Synthetic Molecules Technical Development, F. Hoffman La-Roche AG Basel CH-4070, Switzerland
Abstract
The implementation of Continuous Manufacturing (CM) in the pharmaceutical industry is not advancing as fast as foreseen in its initial years, when the flexibility of this technology encouraged many pharmaceutical companies to evaluate its implementation for commercial manufacturing.
Current approaches try to reduce the gaps in the process control levels for CM with the implementation of process models, e.g., digital twins, and/or reducing the process complexity and thereby the need for a high-level of process control. A particularly promising direction is semi-continuous manufacturing, a hybrid combining most of the advantages of batch manufacturing and CM. Within this space, Mini Batch continuous Direct Compression technology has appeared as an alternative for continuous Direct Compression (cDC), minimizing its downsides by providing a more easily controllable process. This article aims to describe such a technology and its proposed control strategy: Mini-Batch continuous Direct Compression.
Keywords: Mini-Batch cDC, continuous direct compression (cDC), Continuous manufacturing (CM), semi-continuous manufacturing, process analytical technology (PAT)
Introduction
For many years, the majority of solid pharmaceutical drug products have been manufactured batchwise. In batch technology, all formulation components are dispensed into the equipment, and the final product is discharged once the process is completed. The unit operations are performed sequentially one after the other, and the size of the used equipment defines the batch size. The manufacturing time for a batch can vary depending on the complexity of the unit operations as well as the distance between unit operations. The batch size must be defined in the registration dossier, and any change on the batch size must go through a post-approval change. Batches of material that do not meet the pre-defined quality expectations, typically assessed by off-line analyses, are discarded.
By contrast, CM is a rather new advancement in the pharmaceutical industry. This technology offers the potential to increase flexibility, agility and efficiency during development, while delivering a simple and robust manufacturing process with enhanced process control and a flexible batch size.1-3 By 2018, seven continuously produced solid drug products had been approved.4
In fully continuous manufacturing, all unit operations are interconnected to each other: drug substance and excipients are all typically charged at the beginning of the line and the final drug product is discharged at the end of the line. Therefore, the gap time between different unit operations is eliminated due to the continuous transfer of the material through the line.
Mini-Batch continuous Direct Compression (MB cDC)
Direct Compression
In Direct Compression (DC), API and excipients are blended without any intermediate granulation step. DC processes are thus simpler to develop. Historically, batch DC has been applied when the API and excipient properties enabled homogenous free-flowing blends facilitating a robust tablet compression process. Only if these material properties cannot be achieved, or maintained for large batch sizes, does a granulation step become necessary. The introduction of CM has given rise to new opportunities, with lower amounts of material processed at the same time reducing the impact of the material properties on the DC performance. This makes DC more amenable to materials with poorer flowability, segregation potential, higher cohesivity and worse compression properties.
continuous Direct Compression (cDC)
The unit operations for cDC and batch DC are the same: feeding, blending and compression. In DC, transfers between unit operations typically mean manual operations and change between manufacturing areas. In cDC, transfers are automated and the whole process train is located in one room and usually benefits from a vertical alignment.
One of the key benefits of cDC is the ability to adapt the line rate to the predicted market volume for a specific product and adjust the run time to the current supply demand; cDC equipment is commercially available which allows it to run at a high line rate (from 50kg/h up to 250kg/h) usually done for very high-volume products5 to low line rates (from 10 to 25kg/h) for medium and low volume products. The main disadvantage of cDC is the time and therefore product consumption (waste) during the start-up and ramp down phase, where typically the process has not yet reached steady-state and therefore the product does not have the desired quality. Such waste has to be diverted, adversely impacting the yield, especially when the volumes to be produced are small. This is not only relevant for commercial manufacturing but also in development, where the API availability is limited. The low availability of API in development is accentuated due to the current trend of increasing the number of targeted therapies with smaller patient populations, resulting in lower-volume demands.
Mini-Batch continuous Direct Compression
To achieve the benefits of continuous technologies yet minimize its drawbacks, Roche is implementing a Mini-Batch cDC, a semi-continuous manufacturing process. As depicted in Figure 1, it consists of small-scale batch (1 kg approximately) feeding and blending operations, carried out repeatedly at a frequency allowing continuous compression. The synchronized combination of a feeding/dosing time of 1 kg in two minutes, a two minute blend time, and a two minute period to discharge and transfer to the tablet press, produces 1 kg tablets in six minutes, resulting in a line rate of 10 kg/h.
The specific benefits of MB DC have been summarized in,6 and can be articulated along three directions: increased speed to patients, more efficient operations, and more robust processes. The fundamental levers leading to these improvements are the like-to-like transfer to commercial, the simplicity of the process, the flexibility given to the batch size, and the enhanced process control.
In particular, the following steps are highlighted to explain the technology:
Accurate feeding due to discrete dispensing: Feeding is a critical unit operation for technologies with continuous elements.7 In the Mini-Batch technology, the feeders always operate based on the balance signal, which is referred to as gravimetric mode. The feeder refi ll takes place during the blending phase, and the feeders are tared before the feeding takes place. Therefore, changes due to densification or changes of API and excipient lots, will not impact the precision of feeding. The weight dispensed by the feeders directly corresponds to the mass of each component in that particular Mini-Batch and very accurately reflects the quantitative individual Mini-Batch composition (assay). In case the feeder(s) do not meet the target precision, the complete Mini-Batch will be diverted to waste before the tablet press. In continuous feeding in cDC, by contrast, during the refill the feeders have to operate based on the volumetric rate, referred to as volumetric feeding, because no balance data of the feeders is recorded during refill. This is especially critical, since refill potentially causes densification of the powder bed. Therefore, the verification of the Blend Uniformity (BU) (mainly PAT in combination with RTD models)8 is crucial to ensure material diversion in case of inaccurate feeding during the refill.
Verification of Mini-Batch mass: Once the blending process is finished, the Mini-Batch is emptied from the blender. A weight verification takes place in the Gain-in-Weight (GIW) hopper to ensure that all material that was fed in the blender progresses to the tablet press.
No process scale-up: As further outlined in the next section and Figure 2, all equipment is identical during development, transfer and commercial manufacturing. Therefore, scale-up and “change of equipment” effects are excluded, yielding robustness and saving resources. This is especially attractive for the blending process, where scale-up issues most frequently occur. For the Mini-Batch technology, blending is performed on the Mini-Batch scale (~ 1 kg) using the same blender (5L mixer) for all operations. This is an intense mixer, in which the API distribution is expected to be superior compared to a large bin blender. In addition, since no large powder masses need to be moved, the flow requirements for the resulting blends are not as demanding as for a batch process. The potential critical blending process parameters are: blending time and speed. These factors, as well as the critical material attributes for blending, are addressed in process Design of Experiment (DoE). A line rate of 10 kg/h also results in a relative low rotation speed of the press, which is again beneficial in terms of robustness of the compression process even with less favorable powder blend properties. All manufactured and conforming tablet cores are collected and transferred to a coater, where coating is performed as an independent unit operation.
Mini-Batch Process Considerations
Process Configurations
Figure 2 shows the three process configurations to be used at different stages of the drug product lifecycle. The stand-alone “unit operations” configuration allows formulation development and assessment of feeding and blending performance directly at representative Mini-Batch scale and with the identical unit operations. The “integrated line 1” configuration will be used for process development, manufacturing of clinical trial material and primary stability batches at the development facility. The same process configuration will be used for commercial manufacturing (integrated line 2). The design and process conditions of the “unit operations” and the “integrated line” configurations are intended to be identical, with the same feeders, blender and tablet press being used. This way, no or almost no technical transfer activities are expected from process development to commercial manufacturing, significantly reducing overall time to market.
Mini-Batch Intermixing
Intermixing, in the context of the Mini-Batch technology, is defined as the phenomenon by which two or more adjacent Mini-Batches are mixed in the inlet chute and feed frame of the tablet press during the tablet compression operation. Intermixing, therefore, affects the individual Mini-Batch concentration as well as the Content Uniformity (CU) upon tableting.
The level of intermixing was investigated by discrete addition of Mini-Batches onto a conventional rotary press equipped with a standard feed frame. A placebo formulation was used, with tartaric acid as a tracer at three concentrations: 2%, 5%, and 10%. A step profile was used to measure Residence Time Distribution (RTD) curves. The timing of sampling was based on mass throughput. All tablets were measured at line with a transmission Raman spectrometer; a Partial Least Squares (PLS) model was developed to predict the tartaric acid concentration.
A flowsheet model comprising the feeding inlet and the tablet press was developed and the experimental data were used to calibrate the RTD curve. The intermixing experiment was simulated and the resulting curve compared to the experimental results.
Figure 3 shows the comparison of the measured tartaric acid concentration predicted by Raman (yellow symbols in the graph), with the Tablet Assay (TA) concentration curves simulated with the flowsheet model. The Raman predictions show that the concentration transition between Mini-Batches with different tartaric acid concentrations does not follow a sharp step function, indicating intermixing between Mini-Batches. This behavior was also confirmed by the simulations, which yielded profiles comparable (RMSE: 0.75%) to the Raman predictions of tartaric acid in the tablets.
The experimental results and simulations demonstrated that significant intermixing occurs in the tablet press equipped with the standard feeding pipe and feed frame. Quantifying the amount of intermixing is an important part of the control strategy to assure material traceability in the continuous part of the line. The results above led Roche to a control strategy independent from the level of intermixing, and hence implementing a lean control strategy where the tablet content uniformity will be measured and confirmed as IPC.
Mini-Batch cDC Control Strategy
Figure 4 shows the control strategy and its elements for drug product process development and commercial manufacturing. Four different Critical Quality Attributes (CQAs) have been identified for the Mini-Batch cDC manufacturing line:
- Blend Assay (BA): the Mini-Batch composition (%) in terms of API will be calculated for each Mini-Batch after the blending operation. It uses the Loss In Weight (LIW) data from the feeders and compares it against the data from the GIW hopper.
- Blend Uniformity (BU): The API uniformity after blending will be ensured by employing a product specific design space for each Mini-Batch after the blending operation.
- Tablet Weight (TW): TW will be monitored as an In-Process Control (IPC) during the tablet compression operation by measuring the compression force of individual tablets as an indicator of the TW. Tablets that are outside of pre-defined compression force limits will be ejected.
- Tablet Content Uniformity (TCU): The API uniformity at tablet core level will be controlled as an IPC during the tablet compression operation. The frequency of sampling will depend on supporting data obtained during development, project specific properties (API concentration) and the commercial lifecycle stage.
Release testing will be performed by applying conventional end product testing according to ICH Q6A.9 Real-time release testing will be considered for future introduction after the company experience with the technology has grown further.
As shown in the Figure 4 the proposed control strategy has two material diversion points to avoid that non-conforming material is incorporated in the finished drug product:
- Diversion point 1: Before a given Mini-Batch moves to the tablet press, the control system will verify that it meets the requirements for BA, BU and residual mass. This will be done immediately after discharging the Mini-Batch from the blender, in the GIW hopper: if BA, BU and residual mass are within the established action limits, the Mini-Batch will progress to the tablet press. Otherwise, the control system will direct the entire Mini-Batch to waste/quarantine.
- Diversion point 2: rejection of individual tablets during compression based on the main compression force limits.
Summary and Conclusions
Continuous manufacturing offers several advantages compared to batch manufacturing, yet the number of CM applications is still limited due to various challenges, in particular feeding variability during the feeders’ refill, extensive material consumption to reach a state of control, and complex control strategies. In response to those challenges, semi-continuous manufacturing associates the benefits of both continuous and batch manufacturing. Its application to DC, Mini-Batch cDC technology, offers an attractive, optimized combination: low complexity of unit operations, simple process dynamics, API saving in development and high yield. With Mini-Batch DC the commercial process can be locked earlier in development, without the need for scale-up, and also the control strategy can be very lean. Following its strong belief in the Mini-Batch cDC technology described here, Roche is currently intensifying its development and deployment of the technology.
References
- Leuenberger H. New trends in the production of pharmaceutical granules: batch versus continuous processing. European Journal of Pharmaceutics and Biopharmaceutics 52 (2001) 289–296.
- Wahlich J. Review: Continuous Manufacturing of Small Molecule Solid Oral Dosage Forms. Pharmaceutics 13 (2021), 1311.
- ICH guideline Q13 on continuous manufacturing of drug substances and drug products (draft). ICH (2021).
- Vanhoorne V, Vervaet C. Recent progress in continuous manufacturing of oral solid dosage forms. Int. J. Pharm. 579 (2020) 119194.
- Holman J, Tantuccio A, Palmer J, van Doninck T, Meyer R. A very boring 120 h: 15million tablets under a continuous state of control. Powder Technology 382 (2021) 208–231.
- Warman M, Piccione PM, Maurer R. Leveraging a mini-batch wise continuous direct compression (CDC) approach to optimize effi ciency in process development, on-demand manufacturing and continuous process verifi cation (CPV) in Continuous Pharmaceutical Processing and Process Analytical Technology, 2022.
- Hörmann-Kincses T, Beretta M, Kruisz J, Stauff er F, Birk G, Piccione PM, Holman J, Khinast JG. A Novel Holistic Workfl ow for Predicting Feeder- and Feeding-Performance Based on Selected Material Attributes. Int. J. Pharma (2022), submitted.
- Almaya A, De Belder L, Meyer R, Nagapudi K, Homer Lin HR, Leavesley I, Jayanth J, Bajwa G, DiNunzio J, Tantuccio A, Blackwood D, Abebe A. Control Strategies for Drug Product Continuous Direct Compression- State of Control, Product Collection Strategies, and Startup/Shutdown Operations for the Production of Clinical Trial Materials and Commercial Products. Journal of Pharmaceutical Sciences 106 (2017) 930-943.
- ICH Guideline Q6A. Specifi cations: Test Procedures and Acceptance Criteria for New Drug Substances and New Drug Products: Chemical Substances. ICH (2000).
Author Biography
Dr. Manel Bautista received a PhD in analytical chemistry from the Autonomous University of Barcelona in 2009, focused on the use of PAT in the pharmaceutical fi eld. After several years working as PAT expert in Novartis implementing Real Time Release Testing (RTRT) for Oral Solid Forms he joined Roche in 2020 where he supports the control strategy and PAT elements for the drug product continuous manufacturing line.
Reto Maurer is a principal formulation scientist in Formulation R&D of F. Hoff mann- La Roche in Basel, Switzerland. In his role he is responsible for the development and clinical manufacturing of solid oral dosage forms. Besides formulation development, he is leading the drug product continuous manufacturing implementation team in Basel.
Dr. Laura Rolinger studied chemical engineering at Karlsruhe Institute of Technology and the University College London with an exchange to the University of South Australia. She defended her doctoral thesis successfully in 2021. During her PhD studies at KIT, she focused on implementing Process Analytical Technology (PAT) for the downstream process of biologics. Laura joined Roche in July 2020, where she supports synthetic molecule drug product development and the implementation of the drug product continuous manufacturing line.
Dr. Emmanuela Gavi is a principal modeling and simulation scientist in Pharmaceutical R&D of F. Hoff mann- La Roche in Basel, Switzerland. In her role she is spearheading the implementation of process models to support drug product development and is responsible for leading the modeling eff orts accompanying the introduction of the drug product continuous manufacturing line. She holds a PhD in Chemical Engineering from Politecnico di Torino and joined Roche in 2013.
Dr. Patrick M. Piccione is the head of Formulation and Process Sciences at F. Hoff mann- La Roche in Basel, Switzerland. He spent seven years in materials development at Arkema, followed by nine years at Syngenta, leading process engineering science. In his role at Roche, he ensures enabling sciences (process modeling, biopharmaceutical sciences) support towards the development and clinical trial manufacturing of solid oral dosage forms. In addition, he acts as the sponsor of Roche’s drug product continuous manufacturing project.
Subscribe to our e-Newsletters
Stay up to date with the latest news, articles, and events. Plus, get special
offers from American Pharmaceutical Review delivered to your inbox!
Sign up now!