OR WAIT 15 SECS
Patricia Van Arnum was executive editor of Pharmaceutical Technology.
Fueled by a need to reduce costs and improve efficiencies, continuous processing may be the next paradigm shift in pharmaceutical manufacturing.
Continuous processing is taking root in the pharmaceutical industry. Faced with greater pressure to reduce manufacturing costs, the lure of reduced process variability, and a regulatory environment supportive of manufacturing modernization, companies are considering continuous processing in their short- and long-term strategies.
(POWDER: ALICE EDWARD/GETTY IMAGES, ILLUSTRATION: MMCEVOY)
A regulatory framework
Although common in other process manufacturing industries (e.g., food and chemicals), continuous processing is at a nascent stage in the pharmaceutical industry, where batch manufacturing is the prevailing mode.
"The movement from batch to continuous processing may not be as extreme as it seems at first," says Moheb Nasr, director of the Office of New Drug Quality Assessment, Center for Drug Evaluation and Research at the US Food and Drug Administration and member of Pharmaceutical Technology's editorial advisory board. "Many current pharmaceutical operations such as roller compaction and tablet compression can be considered semicontinuous operations and readily could be converted to true continuous mode. These unit operations have a constant flow of material in and out (like a continuous process), but are typically run over a defined period of time or for a fixed amount of material (like a batch process). Other unit operations such as blending, drying, and tablet coating have traditionally been operated in batch mode and could require more development efforts to transition to continuous mode."
By its very nature, continuous processing lends itself to in-process monitoring and control and is compatible with FDA's process analytical technologies (PAT) initiative and the agency's overall move to a risk- and science-based approach to pharmaceutical manufacturing and quality-by-design principles. In its PAT guidance, FDA identified "facilitating continuous processing to improve efficiency and manage variability" such as through small-scale equipment that eliminates scale-up issues, as a way to improve quality, safety, and efficiency (1).
Current regulations do not distinguish between batch and continuous manufacturing. "A point of confusion is the word 'batch,' which can mean either the mode of manufacturing or the quantity of material being processed," says Nasr. The regulations specify (2):
A batch means a specific quantity of a drug or other material that is intended to have uniform character and quality, within specified acceptance limits, and is produced according to a single manufacturing order during the same cycle of manufacture.
"The definition of batch here refers to the quantity of material and does not specify the mode of manufacture," says Nasr. "Manufacturers should not be dissuaded from applying continuous processing based on the prevalence of the word 'batch' in the regulations."
A major difference, however, between batch and continuous processes is determination of product or intermediate uniformity. "In batch processes, uniformity is often determined by measuring or sampling at different spatial positions within equipment; in continuous processing, uniformity would likely be determined across time," explains Nasr. "A 'batch' would need to be defined by the manufacturer for continuous processing that is consistent with the regulatory definition under 21 CFR 210.3(b)," he says. These regulations deal with current good manufacturing practices (2). The European Medicines Agency (EMEA) does not distinguish between batch and continuous processing, according to members of EMEA's Inspections sector.
Pros and cons
Given a regulatory framework for continuous processing, the key question is whether it offers technical and economic benefits compared with batch manufacturing. On that point, there is mixed opinion.
"Selection of continuous processing over batch manufacturing is driven by two factors: economics and process control," explains Anthony J. Maddaluna, vice-president and team leader of Pfizer Global Manufacturing strategy and supply network transformation. "It is expected that a continuous process will use less equipment, sit in a smaller building footprint, use less labor, use less utilities, generate less waste, and hence, will be more economical. A general rule of thumb is that a well-designed continuous plant should cost about 40% less than a comparable batch plant," he says.
Additional factors need to be considered in evaluating potential economic gains. "Given the fact that many companies find themselves with excess capacity, the economic justification for a switch from batch to continuous is not straightforward," says Maddaluna. "It involves consideration of the operational benefits, write-off costs for excess capacity as well as strategic issues. The justification for investment in continuous operations is easier for new products or specialized processes where an investment is required anyway. For example, in the case of processes involving very active APIs or highly hazardous material, where a containment investment might be required, a fully contained continuous process is likely to be more economical than a batch process."
Rationale for continuous processing
Others are more cautious on the benefits of continuous processing. "Until we are able to develop the science and technology to design reliable systems for continuous processing, it will be very difficult to say with certainty that continuous processing will improve manufacturing efficiency or cost," says Prabir K. Basu, executive director of the National Institute for Pharmaceutical Technology and Education (NIPTE). "Today, continuous processes look attractive compared with batch processes because the batch process is not well understood and optimized. There may be a misconception that just by changing to a continuous process all problems will be resolved. But, before one thinks of changing to a continuous process, the question must be asked whether the current batch process is fully understood."
Maddaluna, however, points to potentially achieving better process understanding in a continuous environment. "Process conditions can be more stable through achieving a steady state," he says. "Feedback and feed-forward controls can respond to variations in raw materials. More time and effort is required to design and develop a continuous process, therefore, the process and upset conditions are better understood."
Others agree. "Continuous processing allows quality to be built in the process, to measure in-line, and to adjust parameters to drive the critical quality attributes to the requested target levels," says Wim Van der Goten, sales director for GEA Pharma Systems Collette (Wommelgem, Belgium), a solid-dosage equipment manufacturer. "Current batch production techniques are very often inefficient and cannot manage variability of input material characteristics in a proper way. Therefore, batch processes often have reduced reliability and poor yields. A statistically disputable sampling system after every batch process step decides whether the product can be released for the next production step."
Testing and product release throughout the various steps in a batch manufacturing offer certain advantages. "In a batch process, since the batch is tested and validated, the amount of material at risk is only one blender full of material," says Maddaluna. "For a continuous process, the amount of material potentially at risk is the amount used during the length of time the processes run (i.e., hours, days, or weeks). This concern will likely be alleviated with the advent of PAT and process controllers."
Continuous processing also presents challenges in product homogeneity because of start-ups and shutdowns caused by equipment failures or other operational issues. "It is hard to establish criteria for when to start accepting the product and when to start rejecting. It is very hard to validate such processes," says Basu. "One of the NIPTE members has specific experience in a commercial pharmaceutical process where a semicontinuous filling process was shut down by FDA because of too many starts and stops. The process, equipment, and process-control systems should be very well designed, extremely reliable, and should produce consistent product to avoid frequent shutdowns for continuous processes to be successful."
Basu outlines other limitations of continuous processing for raw materials variation. "While a continuous process is designed to take care of small variations in the raw materials, it is difficult to design ahead of time and predict and control all types of variations and wide swings in the variation of raw materials," he says. "In such instances, a continuous process is more vulnerable than a batch process, where, for example, a PAT tool could be used and variables such as time of mixing can be adjusted to accommodate these types of variations. The raw materials need to be well characterized and variability understood, so that a robust continuous process can be designed."
Continuous processing at work
As economic and technical concerns are considered, companies are advancing projects in continuous processing.
Novartis. Novartis Pharma AG (Basel, Switzerland) is proceeding with what it terms its "Blue Sky Vision" for continuous processing, in which process steps are reduced to a minimum, and a product is made from start to finish, from drug substance to finished drug-product, on a continuous basis in one facility. Novartis has teamed with the Massachusetts Institute of Technology (MIT) to help realize that goal. As part of a 10-year collaborative effort, Novartis is investing $65 million in the Novartis-MIT Center for Continuous Manufacturing, which was formed in September 2007.
"We have a pretty ambitious view on continuous manufacturing," says Walter Bisson, global program manager of technical operations at Novartis. The company's focus on continuous processing is not to simply convert existing processes from batch to continuous, as this would be "a very traditional approach," he says. "We're trying to invent new methodologies, which lead to new technologies, which can then be translated into new equipment." Equipment manufacturers have been involved since the early stages of the group's research.
Continuous processing would further involve integrating separate process steps into one process. Bisson offers an example of reducing the throughput time, or the time it takes to go from a chemical intermediate to the final drug product. Chemical and drug-product manufacturing usually occurs at different facilities, locations, and countries, and the entire process can take 300-plus days, he explains. With a continuous process, non-value-added steps such as transportation could be eliminated because all the steps are performed at one facility. "In the Blue Sky Vision, we can envisage a 10-day throughput time—even if it turns out to be 30 days, this would be a tremendous improvement over what it is today."
He says that although the industry has invested heavily in the past 30 years in improving processes, reducing throughput times, and increasing operational efficiencies, it has always done with batch thinking. "We believe the quantum-leap improvements with the batch concept have been done in the past. Looking forward, we will have improvements, but they will be marginal—probably 3% here or 5% there," says Bisson. "With the improvements we have in mind, we really want to take quantum leaps where we improve by 30%, 40% or more. It sounds aggressive, but yes, our program is ambitious," he says. "If you envisage a 50-day throughput time, or even a 10-day throughput time, just calculate the reduction."
Bernhardt L. Trout, director of the Novartis-MIT Center for Continuous Manufacturing and professor in MIT's Department of Chemical Engineering, explains that the goal of the partnership is to develop a portfolio of technologies to allow the implementation of this integrative approach. Eleven research activities are underway at the center, which involve chemical synthesis, new reactor approaches, including microreactors, and new approaches for separations, crystallization, and final finishing. Forty students and postdoctorate associates, along with 10 professors from MIT's chemical engineering, chemistry, and mechanical engineering departments, are working at the center.
The MIT team aims to have a full-end, bench-scale unit for continuous processing within the next two years. The bench-unit will be scaled up into a pilot-scale unit and transferred to Novartis for development into commercial scale.
Trout says the partnership's goal of a fully integrated process that incorporates both drug-substance and finished- product manufacturing is "the ultimate lean technology." This new, integrative approach will "make the quantum leap, the big jump, or the paradigm shift," in pharmaceutical manufacturing, he says.
Pfizer. Pfizer (New York) is evaluating continuous processing for drug-substance, biologics, and drug-product manufacturing, says Maddaluna. Its manufacturing division is working with its research and development group for:
C-SOPS. The Center for Structured Organic Particulate Systems (C-SOPS), a multi-university consortium consisting of Rutgers University, Purdue University, the New Jersey Institute of Technology, and the University of Puerto Rico at Mayaguez, is developing a test bed for continuous manufacturing. C-SOPS was founded in 2006 with a $15-million grant from the National Science Foundation and is also funded by industrial partners, which include pharmaceutical manufacturers and equipment producers.
C-SOPS is focused on three areas: manufacturing science, composite synthesis and characterization, and particle formation and functionalization. The center is developing a test bed to show the feasibility of continuous technology for sequentially blending, dry-granulating, lubricating, and tableting of dry powders and granules. By using continuous manufacturing methods, this test bed seeks to develop predictive models for single-process components as well as the integration of these models into hybrid platforms to be used for process control, explains Fernando Muzzio, director of C-SOPS and member of Pharmaceutical Technology's editorial advisory board. The goal of the test bed is to mitigate the three most common problems affecting the quality of finished products made by batch processing: segregation, agglomeration, and compact quality while simultaneously improving process controllability and robustness. Other advantages are facilitation of process scale-up and reducing the size and cost of equipment.
Solid-dosage equipment design
Modifying equipment for smaller scale is key for advancing continuous processing. To that end, GEA Pharma Systems has recently commercialized "CONSIGMA," a high-shear granulation and drying system for continuous processing (see Figure 1). The system has three modules: a wet high-shear granulation module, a segmented dryer module, and an evaluation module.
Figure 1: A continuous high-shear granulation and drying system ("ConsiGma," GEA Pharma Systems). (FIGURE 1 IS COURTESY OF GEA PHARMA SYSTEMS.)
In the granulation module, dry ingredients are dosed individually or premixed in the continuous high-shear granulator. After a small dry-mixing section, the granulation liquid is added, so each particle receives the same amount of liquid. The particles follow a granulation track, which mimics the granulation in a batch process. Narrow tolerances between granulation screws and the barrel minimize back-mixing. "The whole wet-granulation process takes place in a few seconds with only a few grams of product in process at the same time, resulting in faster start-up and no losses," says GEA's Van der Goten. The particle size can be adjusted by changing the working level in the granulator, which results in a continuous flow of wet granules with a constant quality and density that is transferred to the dryer. There are no oversized agglomerates and thus no wet-milling.
The dryer module, based on fluid-bed drying principles, splits the continuous flow of granules in packages of 1.5 kg, drying them each in a separate segment of the dryer. When the content of the segment is dry, it is emptied and transferred to the evaluation mode and refilled with a new package of wet granules. The drying curve of each package is monitored. In the evaluation mode, the dried granules can be measured for critical quality attributes such as particle-size distribution, humidity, and content uniformity. At any time, there are only 6 to 9 kg in process, which minimizes the amount at risk in case of an incident (e.g., a power failure). The unit can handle capacities from 500 g to tons, so there is no need for scale-up. Van der Goten says that the unit's small size and modular construction allows for fast deployment and makes it easy to install with existing equipment. "It can perfectly operate in a mixed environment with the possibility to do a parametric release to the next step, a tableting process for instance."
GEA Pharma Systems has sold five CONSIGMA units: two to European pharmaceutical firms, one to an American pharmaceutical company, and two units to two Mexican generics companies.
GEA Pharma Systems also has developed a continuous blender, which can be used for premixing or mixing the external phase into the granules on a continuous basis. Courtoy, another division of GEA Pharma Systems, developed an advanced on-line PAT for its tablet presses, which allows GEA to provide a full continuous production line of solid-dosage forms, says Van der Goten. Such a line is currently under construction at GEA's laboratories in Belgium.
For insight on continuous sterile manufacturing, see "The Potential of Continuous Sterile Manufacturing".
1. FDA, Guidance for Industry—PAT—A Framework for Innovative Pharmaceutical Development, Manufacturing, and Quality Assurance (Rockville, MD, Sept. 2004).
2. Code of Federal Regulations, Title 21, Food and Drugs Part 210.3(b) (Washington, DC, 2008), p. 138.