Cynthia A. Challener is a contributing editor to Pharmaceutical Technology.

Cynthia A. Challener

The pharmaceutical industry is no stranger to fermentation. Synthetic biology has been used for the production of both biologic and chemical APIs for decades. While it has not been the predominant method in either sector, the greater sustainability and ability to access novel structures provided by fermentation processes are driving greater interest, particularly for small molecules.

Synthetic biology “will propel companies that understand the right strategy for the distinct value propositions it offers at each step in the production life cycle,” said Gihan Hewage, previously a Lux Research analyst, in a press release (1). “Companies with products in the ideation and early development stages will differentiate most from flexible production and the ability to create novel products, whereas products in later stages of development can leverage the environmental and marketing benefits of the technology,” he added.

Cell culture, while often more suited for the production of biologics, such as proteins and monoclonal antibodies, is more expensive and often more complex. Microbial fermentation tends to be, according to Jose Luis Barredo, biotechnology business director with AMRI, more straightforward and scalable. One key reason is that the organisms have faster and more stable growth patterns and lower intrinsic metabolic burden, notes Filippo Giancarlo Martinelli, head of global industrial partnerships development at Abolis.

In addition, fermentation often involves the conversion of cheap and readily available raw materials (sugar or other cheap substrates) into high-value products with high yields/titers, ensuring the commercially viable manufacture of high-quality products, observes Nikolay Krumov, senior project leader, upstream development microbial at Lonza.

Processes typically proceed at room temperature and ambient pressure in water, avoiding the use of harmful or toxic solvents or chemical raw materials and production of dangerous waste, according to Martinelli. Operators benefit from safer work environments, and drug makers avoid issues associated with the use and production of hazardous materials. Fermentation facilities and equipment may also be less costly to build and install.

Synthetic biology can in some cases also provide access to molecules that cannot be produced via traditional synthetic organic chemistry. “While this concept is not novel per se, over the last several decades, the development of improved tools and understanding have also enabled greatly increased yields and decreased development costs and risks,” Martinelli states. He points to steroids, analgesics, opioids, and some anti-cancer drugs. Similarly, he observes that some phytochemicals that cannot be practically extracted from plant sources may also be produced via fermentation.

There are also decades of fundamental research and proven manufacturing success supporting the growing interest and application of synthetic biology and fermentation for chemical API production, Krumov asserts. “The expression systems are well known, studied, and characterized, and the processes are well understood and controllable, facilitated by the development and improvement of precise monitoring tools,” he explains.

The key to realizing all of these benefits, however, Martinelli stresses, is to recognize when synthetic biology provides the optimum solutions—either by providing access to novel structures or higher yields/selectivities—and when organic chemistry is the best approach.

AMRI breaks down microorganisms useful for fermentation into two classes: natural, classic producers and engineered host strains (genetically modified organisms). Classic microorganisms widely used in biotechnology processes include bacteria (

), make up the second group, according to Barredo. “These host strains are selected based on suitability for genetic modifications and utility to efficiently produce APIs and compounds, including many products not practical via fermentation previously,” says Peter C. Michels, head of global fermentation at AMRI.

The majority of developmental and commercial processes in the pharma industry involve 

), Krumov adds. “E. coli is probably the most studied and well-known microorganism, a sort of work horse for the biotechnology industry. It provides simple and fast strain development, short fermentation times, high cell densities, and double-digit titers. 

 achieves ultra-high cell densities grown on simple and cheap carbon sources and also secretes the product into the culture medium, delivering a highly pure product with decreased recovery efforts,” he explains.

New species, Krumov notes, including such requiring biosafety level 2 handling, are constantly evaluated and investigated, and some find commercial realization.

Advances in biologic and analytical technologies

A primary driver of recent interest in fermentation is advances in molecular biology and synthetic biology. “These advances allow producer strains to be engineered with greater understanding and effectiveness, which is therefore enabling the development of new processes,” Michels states. They can also be adapted and improved using established fermentation process knowledge and infrastructure that has been built over decades.

The use of synthetic biology, including directed evolution of biosynthetic pathways, offers new, more specific ways of modifying pharmaceutical production, according to Michels. “These new innovations also allow us to get more out of classical fermentation processes. We’ve seen really dramatic advances in metabolic engineering across a broad spectrum, such as directed evolution, that allow us to simply and systematically evolve the biosynthetic pathways and achieve a myriad of advantages—the overproduction of APIs, reduction of related impurities, shortening the fermentation process and making it more cost effective, producing new compounds, improving process safety, and more,” he comments.

Metabolic engineering, therefore, is playing a much greater role because it can be applied to many different goals for continuously improving practical processes. Martinelli agrees that metabolic engineering and genetic engineering play a key role in the development of novel biosynthesis processes because they are crucial to yield enhancement as well as diversity generation. “By rewiring the microbial network, it is possible to drastically divert the flow of material through the cell metabolism and thus achieve production of the target molecule at the desired scale and cost,” he says.

For Lonza, genetic engineering has been most important in its efforts to develop processes for the production of biotherapeutics. “Technical advances in genome editing, next-generation sequencing, and genomics are increasingly important to support the design of productive and scalable production strains,” asserts Joachim Klein, head of strain development and cell banking for Lonza’s microbial development services.

As well as the development and selection of high-performing, stable, and productive strains, high-quality and high-throughput screening—in combination with the development of effective analytical methods for assessing product quality, such as such as post-translational modifications of candidates generated using candidates identified in the screening process—have also been important contributors to the greater application of fermentation in API manufacturing, according to Klein.

Advances in analytical technology, particularly mass spectrometry, nuclear magnetic resonance (NMR), and other analytical techniques that increasingly help to elucidate, in much greater detail, exactly what is going on inside the organisms, have a significant impact in driving much greater understanding and the more efficient design of practical production processes, Michels agrees.

Advances in biologic and analytical technologies have occurred in parallel with similar advances in informatics and other digital technologies. Dual development has been necessary to efficiently process the large quantities of data produced and ensure data integrity.

“Informatics and digital technologies are valuable in genomics, metabolomics, and transcriptomics analyses. They help us to know ‘

’ the modifications in the genome of the microorganisms necessary for the construction of an improved strain,” Barredo states. Special software, Krumov adds, are applied in the elaboration of process models, statistical analysis and predictions, parameter screening, and process optimization, and to establish more precise and agile process control.

In addition, with the growing complexity of target molecules/metabolites to be produced, it is becoming more difficult to design and implement a full genetic engineering campaign and the corresponding large-scale metabolic network rewiring needed to achieve the target yield, according to Martinelli. “Bioinformatics plays a key role in not only the discovery of the relevant enzymatic and cellular regulatory steps, but also in the design of the cloning tools required to implement such changes and the dosing and fine tuning of the changes needed, taking into account the implications of the different omics disciplines, such as transcriptomics, proteomics, metabolomics, and fluxomics,” he asserts.

“Without digital technologies,” Martinelli continues, “it would not be possible to operate at the current level of complexity at the interface between expert scientists and robots.” In fact, to make retro-biosynthesis more feasible, Abolis has developed a proprietary computer algorithm in which artificial intelligence is applied to bioinformatics for gene and pathway selection and manipulation towards the biosynthesis of target molecules.

Advances in process analytical and control technologies are also enabling improved fermentation processes. For instance, new in-process controls (IPCs) are being developed to control the evolution of raw materials, enzymatic activities, product, etc., according to Barredo.

“In addition to more traditional temperature, pH, dissolved-oxygen, carbon dioxide, and oxygen-demand sensors, biosensors are helping to control the fermentation process, and monitoring systems for submerged fermentations have taken a considerable step forward,” Barredo says. He also stresses that the availability of reliable sensors that are able to provide real-time, non-sampling-based process monitoring of physiologically relevant parameters such as cell mass, substrate, and metabolite concentration allows for automatic control of fermentation processes based on suitable process models and control strategies.

In particular, Michels notes AMRI has found a much broader array of practical approaches using NMR, in particular development insights and quality control, and most notably in combination with the use of in-process controls and mass spectrometry.

While much of the substantial capital investments made in the pharmaceutical industry for classical fermentation can be adapted for new synthetic biology processes, there are some new developments in equipment worth noting. Krumov points to new fermentation system concepts and designs that enable greater automation; intuitive, precise, and advanced parameter programming; and online monitoring of essential quality attributes. “New primary recovery equipment (separators, filters, membranes) also ensures better performance, higher productivity, and greater process robustness, facilitating better purification of the products and higher yields,” he adds.

Mobile applications, according to Michels, allow remote monitoring and control of fermentation parameters even from a smart phone and provide much better information, while parallel fermenter systems provide faster and improved development arrays for modified strains and multiple parameters. He also notes that the wider variety of impellers available today afford improved oxygen availability and mass transfer, particularly for the new engineered host strains.

Going forward, the focus will largely be on downstream processing, which according to Martinelli, represents a challenge when scaling up production of small-molecule intermediates and APIs produced via fermentation.

All of these rapid advances in molecular biology and engineering tools will impact fermentation in the pharmaceutical industry over the next several years. “We anticipate that continued exponential growth in molecular biology techniques—actually editing and designing new producer strains at lower cost and higher volumes—will have a significant impact on the design of new biosynthetic pathways in a synergistic way with chemical synthesis in the years ahead, not only in pharma, but in food, agriculture, enzymes for biocatalysis, and more” Barredo asserts.

“A longer-term vision,” Barredo adds, “might include using new competitive fermentation processes based on the circular economy, where food wastes are used as raw materials to design new, sustainable processes that contribute to the safety and quality of the global environment.”

Modern fermentation process development is, according to Krumov, a long, profound process that leverages the knowledge and experience of generations of bioprocess engineers in combination with current cutting-edge technologies and digital tools, to deliver the safest and most effective next-generation medicines. “The steadily increasing worldwide demand for microbial fermentation production capacities is an indicator of the qualities of this biosynthetic approach and recognition of its potential,” he asserts. Martinelli adds that this technology not only brings innovation to create new entities and new treatments, but to limit the environmental impact of their production.

One way to reduce development timelines, Krumov says, is through intensification. “For instance, intensification of process development will reduce the necessary gene-to-vial time through application of current design-of-experiment approaches, soft sensors, and data-based process modeling,” he says. Ensuring and confirming a product’s quality at the very early project stages will also be important, according to Krumov. Where possible, avoiding the use of complex compounds will further facilitate the development of robust, reproducible, and cost-effective fermentation-based manufacturing solutions.

1. Lux Research, “Lux Research Predicts Synthetic Biology Has High Potential as an Alternative Production Method for Chemicals,” Press Release, April 2, 2020.

Cynthia A. Challener, PhD, is a contributing editor to 


Vol. 44, No. 11
November 2020
Pages: 24–28

When referring to this article, please cite it as C. Challener, “

Fermentation Finds Fans in Small-Molecule API Synthesis

Articles

Biological and analytical advances enable modern fermentation processes to deliver safe and effective next-generation medicines.

Excipients are widely recognized as vital to the development of effective small-molecule and biologic drugs, with a wide range of approved compounds of varying chemistries providing functionality from taste-masking to stabilization. The latter is also crucial for vaccines and live biotherapeutic products (LBPs), which can suffer degradation during both upstream and downstream processing, fill/finish, packaging, storage, and transport. The properties of different LBPs and vaccines can differ widely, and thus careful selection of the right excipients is essential. Doing so is not easy, however, and requires experience with these challenging products and access to effective screening technologies.

As understanding of the role the human microbiome plays in human health and disease has increased, many academic research groups, startups, and Big Pharma firms have developed microbiome-based diagnostics and therapies, a number of which are advancing through the clinic. In 2018, at least 200 companies were active in the microbiome space, and 2400 clinical trials were underway (1); between January 2016 and July 2019, more than $5.4 billion was invested in partnerships and acquisitions related to LBPs (2).

Stability challenges with LBPs and vaccines

LBPs by definition comprise living organisms that exhibit a desired activity within the human body. They have, in fact, been incorporated into probiotic foods and dietary supplements for many years, but typically, excess colony forming units (CFUs) are used to compensate for any loss of viability upon storage. That approach isn’t possible for pharmaceutical products, so the goal during manufacture is to maximize the number of cells that are viable throughout the entire production process, according to Aaron Cowley, chief science officer with Arranta Bio.

“One of the key attributes associated with LBPs is the ability to keep the cells alive and healthy over a duration of time, thus allowing them to function as they are designed,” says Synlogic Therapeutics Chief Operating Officer, Tony Awad. The sum of the stresses that the organisms experience during upstream, downstream, fill/finish, storage, and rehydration will determine if the organisms will retain viability and activity, he adds. Each of the processing steps, therefore, needs to be optimized, because they can all have an impact on subsequent stability.

Processes also need to be designed for scalability, Cowley adds. “Because most of the organisms that comprise LBPs tend to be unstable, it is essential to minimize process times in order to minimize die off. Effective small-scale processes that take much longer to complete when performed at large scale are, therefore, not practical for commercial LBP manufacturing,” he explains.

Viability, and thus stability, depends on a number of factors, including not only the general fermentation conditions and harvest technique, but also excipient/cryopreservative selection and the lyophilization process, according to Cowley. He also stresses that development of an optimum cell bank is crucial as well. “Optimizing the master cell bank is necessary to ensure that the organism being produced is optimally healthy, exhibits the right performance behaviors, and is generated in the maximum yield,” he observes.

Whether the strains are delivered as a single organism or consortium can also affect the stability of LBPs, as can the type of organism. “For example, spore-formers can have greater stability allowing for storage at room temperature but may come with side effects as these bacteria have the potential to become pathogenic,” Awad observes. Comprehensive characterization of LBP strains is also essential. “Knowledge of critical parameters and acceptable ranges—redox, osmolality, etc.—and how to control them is necessary to achieve and maintain the proper functionality throughout the production process,” Cowley asserts.

Stability issues can also occur at any stage of a vaccine’s development lifecycle, according to Jee Look, senior director of drug product development for Emergent BioSolutions’ contract development and manufacturing organization (CDMO) business. “In some cases, instability that occurs during upstream/downstream processing can be mitigated during final drug product manufacturing, but it is important to identify and understand the mechanisms of instability early in the pre-formulation stage as instability in the final drug product would have the largest impact on patient safety and overall production efficiency and would be more costly to fix,” he says.

The instability characteristics possible with a given vaccine also depend on the vaccine type, whether it is a live virus, inactivated virus, virus-like particle, or some other type of subunit, recombinant, or conjugate vaccine. The use of aluminum adjuvants can also create different challenges with respect to vaccine stability. “It is important to understand the mechanism of action of each type of vaccine in order to design the formulation and deal with the instability issue,” Look concludes.

Role of excipients in LBP and vaccine stabilization

Stabilization of LBPs involves different mechanisms and thus different excipients. For LBPs stored at frozen conditions, cryoprotectants are necessary, while those that undergo lyophilization will require lyoprotectants, according to Awad. For these applications, excipients are added to the cell slurry obtained following harvest to maintain viability through the harsh freezing and drying process, which can be damaging to live microorganisms, according to Cowley.

“Both of these options can lead to better shelf life of the LBP,” Awad observes. Excipients are also used in final LBP formulations to influence cellular size and structure and help stabilize LBPs during cryopreservation and/or to preferentially bind moisture to ensure viability and activity are maintained once the product is packaged, Cowley adds.

Many organisms developed as live therapeutics also tend to combine into long chains. “These aggregates can lead to low CFU results because many tens of cells (or more) are detected as just one colony,” Cowley explains. In this case, use of appropriate media can minimize aggregation. “At Arranta Bio, we have found that certain media formulations in particular prevent aggregation regardless of the type of organism, allowing for the development of optimum processes,” Cowley says.

For vaccines, different types of excipients play important roles during vaccine production, formulation, fill/finish, and storage processes, according to Look. “Selection of excipients can be a complex consideration beyond stability issues and can also depend on the target product profile (TPP), logistics, and business market complexities,” he comments.

For instance, if the TPP requires a long shelf life and/or high temperature storage, the vaccine would need to be converted into solid form. In this situation, Look says sugar excipients such as sucrose could play an important role. For live vaccines, which can be susceptible to manufacturing and storage loss, some type of protein stabilizer such as human serum albumin or gelatin may be required.

Antibiotics are also used to prevent growth of bacteria during production and storage, and stabilizers such as monosodium glutamate (MSG) and 2-phenoxyethanol are used to protect vaccines from heat, light, acidity, or humidity during processing and storage.

Cryoprotectants and lyoprotectants are the most critical excipients for LBP stabilization and function by minimizing cell damage caused by ice crystal formation during freezing, according to Awad. Some cryopreservatives penetrate the cells, but all generally influence viscosity, osmolality, surface tension, and other properties, according to Cowley. “The goal is to choose the right mixture that provides the optimum conditions for each specific organism to survive during cryopreservation,” he says.

This type of protection, according to Awad, is important whether the LBP is stored as a frozen liquid or lyophilized to produce a powder for oral suspension. He also notes that additional excipients may be required if the mechanism of instability is known. “For example, antioxidants may be utilized if the strain is susceptible to oxidation or buffering excipients may be utilized if maintaining pH is important,” he explains.

The types of excipients used to formulate a specific vaccine will depend on the type of degradation mechanism of that vaccine. “Degradation mechanisms include aggregation, precipitation, oxidation, deamidation, shear, freeze/thaw, air-water interface, thermal, etc.,” Look states. For instance, different protein stabilizers will be required depending on the causes of aggregation, while antioxidants are important to prevent oxidation. As one example, Look notes that a surfactant such as polysorbate 80 is typically used to deal with aggregation/precipitation due to interactions that occur at the air-water interface. “In this situation,” he observes, “polysorbate 80 would migrate to the air-water interface and prevent the protein from getting through.”

Understanding the mechanisms of instability to identify excipients that have been demonstrated to overcome these mechanisms is the best approach to selecting optimum stabilizing agents for both LBPs and vaccines. For many of these products, however, the role excipients play is still unknown. “Trial and error, empirical approaches, and historical experience are needed when considering excipient options, especially for new gene-therapy vaccines,” Look comments.

For LBPs, the effectiveness of excipients is evaluated by testing for key quality attributes or release criteria, including cell counts, viability, activity, and shelf life, according to Awad. “Excipient screening through compatibility studies is also often required,” he adds.

One of the main challenges for LBPs is a lack of commercially available excipients and media blends that have been specifically developed for use with these new active ingredients, Cowley notes. Based on its more than 10 years of working with LBPs, Arranta Bio has addressed this issue with the development of a number of proprietary Arranta Media Blends and CryoPreservatives, both based only on plant-derived components, that the company uses to rapidly screen against client microorganisms to identify optimum process conditions.

“We have tested both the media blends and cryopreservatives against commercially available formulations for GMP manufacturing with good results. Our media blends provide higher yields and more stable organisms, with additional indirect benefits downstream, including higher resulting viability and shorter lag phases through freeze drying. Our cryopreservative blends also afford higher viabilities and in general help us consistently achieve our minimum viability target of 40%,” Cowley says.

For vaccines, the selection of optimum excipients requires a “risk and benefit” analysis, according to Look. “A vaccine designed for healthy babies would have different considerations than a vaccine designed for terminally ill cancer patients,” he notes. Most vaccines are injectables and applied to large healthy populations, therefore selection of optimum excipients is more rigorous and depends on several factors, Look adds.

“The initial selection of excipients and concentration usually starts with historical information such as those already used in commercial vaccines. From there, the final selection and optimal concentration is done through pre-formulation and formulation development studies with the specific vaccine,” he explains.

Awad recommends that for any drug formulation, it is generally best to use excipients that have been included in prior approved products rather than novel excipients, because the groundwork has been laid and these excipients are likely to have regulatory agency approval. If a novel excipient is needed, though, additional testing, such as an extended animal study (tox study), may be needed, according to Look.

It is also important, Look reiterates, that excipients are not the only means to resolve vaccine—or drug—instability problems. “A more holistic approach is needed, which includes the consideration of processing conditions and any device used for administration. Some instability issues can be mitigated through process optimization and selection of the appropriate device,” he concludes.

1. C.R. Fernandez, “No Guts, No Glory: How Microbiome Research is Changing Medicine,” 

, Jan. 22, 2019.
2. A. Guillermo, “Why Investors Trust Their Gut About the Human Microbiome,” 

Cynthia A. Challener, PhD, is a contributing editor to 


Vol. 44, No. 10
October 2020
Pages: 20–23

When referring to this article, please cite it as C. Challener, “Factoring Stability in the Biologic Drug Mix,” 

Excipients must be carefully chosen to ensure optimum protection for vaccines and live biotherapeutic products.

Factoring Stability in the Biologic Drug Mix

Through its Operation Warp Speed program for COVID-19 vaccine manufacturing, the US government is paying manufacturers and contract service providers to scale up the manufacture of specific vaccine candidates. If those specific candidates are not successful in the clinic, those facilities must be able to quickly switch to production of different vaccine candidates that have been approved.

While this approach sounds like a practical approach to accelerating vaccine manufacturing, rapid changeover of biologic production process carries numerous challenges. There are a few different strategies for overcoming these challenges, with key components including facility design, equipment selection, operational considerations, and the use of appropriate plant management systems.

Challenges to rapid changeover come from many sources, including the equipment, room configuration, documentation, people, and environmental monitoring, according to Sebastien Ribault, head of end-to-end solutions at MilliporeSigma.

The traditional method of completely stripping down all equipment and installing dedicated materials and components for each product has proven to be a cumbersome and costly process for multiproduct facilities, adds Piergiuseppe Nestola, senior platform technology consultant at Sartorius. “It is not unusual for changeover to require weeks to complete when traditional methods are followed. With the changing industry tide favoring multiproduct manufacturing facilities, methods to enhance efficiency without sacrificing quality or safety are needed,” he says.

Several activities must be performed in a facility to avoid any risk of cross-contamination between Product A and Product B, in particular via a thorough cleaning-in-place of all equipment surfaces in contact with intermediates and the end product, Thibaud Stoll, global head of commercial biologics operations at Lonza, specifies.

For certain biologics, most notably viral vectors, vaccines based on live viruses, and spore-formers, precautions must be taken to prevent spread through the air and more extensive cleaning protocols are generally required, typically with airborne peroxide methodology, says Katarina Stenklo, enterprise solutions commercial activation leader with Cytiva. Any live biological agents, whether they be the host organisms or adventitious agents, must be effectively removed/inactivated from HVAC, facility, and equipment surfaces (both exterior and product paths) prior to introduction of a subsequent product, adds Syed T. Husain, senior vice president and CDMO business unit head at Emergent BioSolutions.

Stainless-steel equipment specifically presents a common challenge, as it requires cleaning and validation of product contact surfaces before the equipment can be used for a subsequent project, according to Evan Shave, director of biomanufacturing—global network operations for Thermo Fisher Scientific. Cleaning and sterilization of stainless-steel equipment is well understood, but often “overdone” in terms of cleaning and rinse times, mostly due to conservatism or a lack of time to properly optimize the process, adds John Machulski, vice president of engineering for Catalent Biologics.

Changeover of processes performed in a ballroom setting can be even more difficult than for processes run in individual suites due to the added risk of cross-contamination with other processes on top of concerns regarding product carryover, notes Steve Gravallese, enterprise solutions program director at Cytiva.

Stoll adds that some product-dedicated equipment must also be installed, in particular chromatography columns with packed resin for drug substance purification. “All these activities may potentially be quite long and on the critical path if not optimized and sequenced properly, and if the facility layout has not been designed accordingly,” Stoll states.

For drug product manufacturing, it is critical to have a lean process design due to its highly kinetic nature, according to Machulski. “Although barrier isolation technology provides the highest level of aseptic processing, long bio-decontamination cycle times remain one of the biggest challenges for achieving a rapid changeover,” he says. Stoll also notes that given the completely different types of equipment, changeover times are difficult to compare. “A change in vial format may require some significant setup activities for a filling line; the number of such changes can be optimized via the proper yearly planning and sequencing of campaigns,” he observes.

For contract manufacturers in particular, the biggest difficulty is the need for processing equipment flexible enough to accommodate different processes/scales corresponding to molecules from different clients. “This challenge is more likely to apply to drug substance rather than drug product because drug substance processes typically have significantly more variability between different products,” Shave notes.

Strategies for optimization of process changeovers differ because the equipment and room configurations for each process are different, but all are driven by the same constraints and rational, according to Ribault. “A combination of primary (process design), secondary (facility design and operation), and procedural controls and studies should be used,” states Husain.

Stoll identifies three key approaches to optimizing changeover activities: making them as short as possible, performing them during “hidden time” or in parallel, and reducing the number of changeovers by running longer campaigns. “Operational excellence/lean methodologies can be quite powerful for optimizing changeover processes, as can the use of redundant equipment or dedicating packing rooms for chromatography columns. In the latter case, there is additional capital expense, but it is usually a good investment,” he says. Running longer campaigns, though, carries a trade-off with the optimization of product inventory.

The ultimate goal, according to Gravallese, is optimizing the facility around unit operations so they can be segregated where open processing cannot be avoided and staggered changeovers are possible. “It really comes down to identifying the best methodologies and designs to stagger processes in facilities without impacting operations in the balance of the site,” he states.

Properly optimizing cleaning and sterilization cycles where stainless-steel equipment is used, validating procedures, having a well-trained staff, and having a robust yet simplified supply chain are keys to facilitating rapid changeovers, according to Machulski. “Additionally, ready-to-use primary and secondary components can be helpful, along with automation and robotic technology for drug product fill/finish manufacturing,” he says.

Ribault also asserts that it is important to look across the entire value chain and consider how to ensure the interface between process development and manufacturing works well; another consideration is to put together a team that will make the project a success in record time. “You do these things by having templates in place for process development and manufacturing that reduce the transition time. So, while rapid changeover is certainly an integral aspect of operations, it is important to keep in mind that it is one piece within the entire value chain,” he says.

In addition, a quality risk management approach may be applied to assess and continuously improve established changeover processes, according to Nestola. “All processes, including changeovers, can be improved with investment of money and resources, parallel activities, equipment design improvements, and standardization,” he asserts.

Enabling changeovers through facility design

Facility design plays an important role in facilitating rapid changeovers. Inclusion of multiple upstream suites can enable better utilization of downstream operations because upstream processing takes much more time, according to Stenklo.

For a multi-product downstream suite, Shave notes that it can be better to have the plant broken up into two or three segregated smaller rooms versus a single ballroom so changeover can start as soon as the unit operation is finished. “This approach affords the opportunity to increase facility throughput,” he says. A phased approach allows for proper and thorough cleaning as the process completes, agrees Husain. In addition, all surfaces should be smooth to allow for easier yet effective cleaning/disinfection, he notes.

Separate mechanical air handling systems should be used for independent spaces to allow some rooms to be shut down while others continue to operate, adds Gravallese. Door interlocking alarm systems are equally important because they ensure that adjacent process spaces remain separated during changeover/cleaning. The ability to monitor spaces quickly is also essential so they can be released back to operations, he adds. “Environmental monitoring of different rooms is necessary to ensure they stay within qualified requirements (temperature, pressure, humidity, bioburden, etc.). Some sampling is manual, but where possible automation of sampling can help speed up changeovers,” Gravallese says.

Other important facility design aspects for rapid changeover, according to Machulski, include: quality by design, design for manufacturability principles (i.e., lean, single-minute exchange of dies [SMED], etc.), integration of lean operating principles such as 5S for optimal equipment, material and personnel flow in the manufacturing suite, and the establishment of SMED quick changeover principles to ensure efficient reconfigurations of the equipment.

Changeovers can also be facilitated if extra rooms are included in the facility where cleaning-in-place activities on equipment can be performed in parallel, Stoll observes. “It’s not just about one aspect, but all of the aspects, including the size of the corridors, equipment placement in the suites, management of sampling, and environmental monitoring working together,” asserts Ribault.

The so-called “facility of the future,” concludes Nestola, would allow rapid changeovers due to its modular approach and because it is fit by design to accommodate single use and minimize any possible cross-contamination while still allowing detection and prevention of possible deviations.

In general, the equipment developed for the biopharmaceutical industry includes built-in functionality—both hardware and software—to make changeovers easier, according to Stenklo. If the same equipment can be used for multiple different processes to avoid having to schedule time during the changeover to move equipment in and out of storage, then changeovers can be faster, adds Shave.

Closed-system design and operation in and between unit operations also mitigates the potential for cross contamination, states Husain. Closed processing is essential because it provides a better response when using a risk-based approach, adds Nestola.

Single-use equipment is therefore one of the biggest equipment-related solutions for facilitating more rapid changeovers. “Pre-gamma irradiated single-use flow paths for all unit operations, standardized pre-sterilized tubing assemblies to connect unit operations, and aseptic quick connectors (rather than welding or heat sealing) are all designed with features that enable rapid changeover,” Shave observes.

The key advantage of single-use equipment is the elimination of cleaning and cleaning validation steps, which significantly reduces cleaning time, according to Stenklo. “With optimized consumables and consumable assemblies, the equipment can be specifically designed for the process and the facility as well,” she adds.

Single-use equipment allows a changeover in a matter of hours because it involves closed bags, according to Ribault. Both the monitoring and documentation steps are simplified because there is lower risk and the process is less complex with single-use versus stainless steel equipment. “Many people who are trained on single-use equipment find it beneficial due to its simplicity and speed of changeover,” he says. Ribault does add, though, that if a hybrid approach is used, all aspects including equipment and sampling, need to be taken into account to ensure there are no bottlenecks that can slow down the changeover process.

Shave adds that flexible hardware with single-use product-contact flow paths are ideal. For instance, he notes that a tangential-flow filtration skid can be equipped with small and large pumps and the ability to take small and larger single-use tubing sizes to accommodate a wide range of processes.

Emergent’s Baltimore Bayview facility, which is the company’s pandemic response facility designated as a Center for Innovation in Advanced Development and Manufacturing (CIADM) by the US Department of Health and Human Services, uses flexible manufacturing and single-use bioreactors to respond to the COVID-19 pandemic. “Flexible design will help enable us to complete rapid changeover and meet the stringent timelines required for producing COVID-19 candidates,” observes Husain.

Even with conventional stainless-steel equipment, there are steps that can be taken to reduce changeover times. “The ability to isolate the process equipment and transfer piping for clean-in-place (CIP) and steam-in-place (SIP) operations is critical,” asserts Gravallese. “Staggering the CIP/SIP of equipment and transfer lines can also accelerate changeovers by ensuring that the critical equipment can be isolated for priority cleaning to facilitate the start of the next campaign. Once all critical equipment and piping has been released back to operations, all secondary equipment can be cleaned for changeover, without impacting the production schedule,” he notes.

Ready-to-use components used within a filling line also allow for the cleaning step in fill/finish operations to be completely removed, while an automated high-speed filling line provides higher capacities than a traditional filling line. “Both of these design elements theoretically allow for more production runs to occur,” Machulski observes. He adds that high operational availability is achieved by reduced line clearance times and changeover times. For example, by design there are 40% fewer parts to change on an automated high-speed filling line than a traditional vial line, and 20% fewer parts to change than a traditional syringe line. Automated servo motor adjustments and robotic arms are the enablers for this step change and result in a reduction of up to 30 minutes per changeover, according to Machulski.

Implementation of rapid real-time testing for bioburden and endotoxin and total organic carbon can also be an advantageous best practice if equipment cleaning is required during change over, according to Shave.

Don’t forget operational considerations

In addition to facility and equipment design, operational aspects can contribute to more rapid process changeovers. In particular, Stoll points out that automation and digital tools are enabling technologies for streamlining changeovers, such as via optimized scheduling of activities and electronic checks and instructions to minimize the risk of errors. “The right manufacturing execution system (MES) can help your facility of the future be both agile and compliant,” asserts Nestola.

Automation software can provide guidance in ensuring that a single-use flow-path is properly connected, thereby saving time performing manual validation, Shave adds. The software can also generate automated batch records of installation and validation, saving the operator time doing manual paperwork. Augmented reality tools, meanwhile, can speed up the changeover process by walking the operator performing the changeover through the process as it is completed.

When one automation platform is used to integrate various pieces of equipment and provide additional functionalities, the ability to scale the whole system with the technology is facilitated, according to Stenklo. “Release handling is much easier when batch reporting, the data historian, and data analytics are maintained in a central place,” she says. “Process data can be accessed while a process is running without the need to gown up and enter the suite. Similarly, systems have been developed to support more rapid changeover approaches.”

In fact, instead of weeks of review, a facility of the future can achieve real-time release without quarantining the product, notes Nestola. “In addition,” he says, “use of analysis of the data collected by the manufacturing execution system can provide information about trends during different changeovers and their possible impacts on the product, if any.”

It is important, though, notes Husain, that automation be designed to allow for plug-and-play operation. “Generally, islands of automation concepts are used. A flexible data historian design also helps in allowing for equipment flexibility yet proper operational control and data analysis,” he explains.

For final drug product manufacturing, Machulski highlights the short bio-decontamination cycle times achieved through optimized injection of hydrogen peroxide vapor and use of catalytic aeration technology within automated filling lines, which enable filling of thousands of components within minutes to hours depending on the total production run. “This allows for more production runs to happen within a set timeframe. Additionally, these automated lines mitigate the risk of human contact with the product, allowing for less variability in the time taken for a human interaction and hence quicker filling times,” he observes.

Plant management systems bring it all together

It is important to keep in mind, though, that automation and software are just tools, not an objective, per se. “They don’t speed up changeover, but can slow it down. Since a facility is the sum of equipment, automation, and people, etc., it should not be considered separately,” Ribault asserts.

In fact, anticipation, training and processes including the mapping of activities and their management through solid project management procedures can eliminate most of challenges, according to Ribault. Electronic or paper procedures that guide shop floor people through all of the steps are key to ensuring fast and error-free changeovers, particularly when combined with MES, electronic batch record systems, and automation, agrees Stoll.

To facilitate rapid changeovers, an overall plant management system should have an adaptive scheduling system that automatically adjusts in response to changes, Shave adds. Supply chain management to ensure supply of consumables is also important, according to Stenklo.

Specific components of a good plant management system include electronic batch record capability, role-based recipe management that optimizes each stage of a recipe lifecycle, data integrity features to help prevent documentation or human errors, and data analytics to monitor, control, and predict any possible variation in the process, according to Nestola.

Overall, Ribault notes that the most important features in a plant management system are people and expertise. “The expertise to run a plant is not in the systems or equipment, but in the team and its experience, including quality, regulatory, and technical experience,” he asserts.

With COVID-19 and the focus on accelerated development and production of potential vaccines, the industry is entering “brave new territory”, according to Gravallese. “It will be even more critical under these circumstances to rely on past experiences to accelerate the approach,” he says.

Any accelerated strategy that involves pushing the boundaries of the usual regulations should be identified up front and discussed with the regulators as early as possible (e.g., use of stable clone cell bank or transient transfection for good manufacturing practice production), Shave stresses. Vaccine developers also need to ensure flexibility with respect to the expected required manufacturing capacity, according to Stoll. “One approach is to work with contract development and manufacturing organizations (CDMOs) to complement internal capacity,” he observes.

When selecting CDMO partners for accelerated COVID-19 projects, vaccine developers should first conduct a risk analysis to identify their capabilities and any gaps related to the projects, according to Gravallese. “It is rare to find a perfect match, but it is important to find CDMOs with the platforms, technologies, and capacities that require the minimum level of retrofits and downtime and will allow for a reasonable best approach. Where customers lack specific knowledge, CDMOs should be able to make assumptions based on their experience combined with good engineering practice, good quality practice, and the available tools in their toolboxes,” he concludes.

It is also important today for CDMOs to specialize in a couple of vaccine platform technologies, adds Nestola. “Most of the vaccines in development are based on one manufacturing platform and many viral vectors processes are similar to each other. The same goes for the mRNA platform, where processes are similar between different candidates. By using multiple platforms, the risk for failure can be mitigated,” he explains.

In addition, given the current increased lead time of most of the critical consumables, Shave says it is imperative that processes be developed using platform raw materials that can be purchased upfront before the process is finalized. “If a CDMO is involved, there will be a need for the drug developer and the CDMO to partner closely together to balance and adjust processes as needed during these unprecedented times,” he remarks.

CDMOs also need to agree with clients on a rapid approach to document review, according to Shave. As an example, he points to the use of templated documents, which can avoid or significantly minimize the need for client review of every batch record. In addition, quality systems may require some flexibility in the usual practices around document control to account for last-minute changes, Shave adds.

One challenge when entering new territory is the rapid pace at which a larger number of activities are occurring. Standardization can, according to Gravellese, help expedite the process. “While the initial reaction of many people is they don’t have time to fit standardized approaches to their specific processes in such an environment, the truth is that standardization actually provides more flexibility. Even if standardization is only achieved on a microlevel, the focus can then be directed to items that are critical and for which true solutions don’t yet exist,” he explains

To start, facilities with segregated production suites should be selected to enable processes to be up and running quickly, but at some point, optimization should be pursued to increase throughput on the same equipment, according to Stenklo. “Similarly, while standalone equipment might provide the fastest pathway to an engineering run, integration of equipment on a single platform will likely provide more efficiency in the long term,” she elaborates.

The parallel investment in CDMOs and vaccine innovation is unprecedented and critical to being able to rapidly respond to the COVID-19 pandemic, Husain concludes. “The potential success or failure of a vaccine candidate cannot be predicted. It’s important that CDMOs continue to focus on supporting all innovators and have clinical and commercial capabilities to continuously pave the way for innovation and development of life-saving medicines for patients,” he says.

Cynthia A. Challener is a contributing editor to Pharmaceutical Technology

Vol. 44, No. 10
October 2020
Pages: 48–54

When referring to this article, please cite it as C. Challener, “Strategizing for Rapid Changeovers in Biologics Manufacturing,” 

Facility and equipment design are important, but the team and its experience matter most.

Strategizing for Rapid Changeovers in Biologics Manufacturing

Rapid changeover of biologic production processes carries numerous challenges. Challenges to rapid changeover come from many sources, including the equipment, room configuration, documentation, people, and environmental monitoring, according to Sebastien Ribault, head of end-to-end solutions at MilliporeSigma.

Cynthia Challener is a contributing editor to 

Equipment and facility cleaning is crucial, with more extensive protocols needed for some biologics.