New Feature: Process Development and Control for New Modalities
Cynthia A. Challener, BioPharm International, BioPharm International, November 2022, Volume 35, Issue 11, Pages: 10–15
Increasing molecular diversity is creating a need for the reinvention of process development and control strategies.
Increasing molecular diversity leads to more challenges in process development and control. Advancements in technology for manufacturing and online analytical monitoring have not caught up to the innovative drug candidates and emerging modalities present in the candidate pipeline today. Lack of pre-determined process and analytical platforms for these modalities often translates to longer development times. New modalities are more difficult to fully characterize due to their complexity, and some highly complex tests take weeks to complete. Therefore, failures in quality attributes are typically found post-production after testing is completed.
Supply chain constraints on critical raw materials in the form of lead times of months to years can be another serious bottleneck for production. Cold-chain requirements have also limited global distribution, storage, and access especially in regions with less developed infrastructure.
Biopharma manufacturers must find solutions to enable the acceleration of the development of the diverse array of novel drug candidates in the pipeline and thus ensure that safe and effective novel treatment options can be brought to patients more quickly.
Several new modalities are challenging biopharma developers
Rapid evolution in therapeutic modalities is currently underway, with autologous and allogeneic cell therapies, viral vectors, plasmid DNA (pDNA), messenger RNA (mRNA), and oligonucleotide therapies, for example, under development as treatments for infectious diseases and cancer vaccines, some eventually as personalized medicines, and some already approved, according to Anne-Cecile Potmans, general manager, Fast Trak with Cytiva. Bispecific antibodies (bsAbs) also present new challenges concerning process development and control, adds Thaddeus Webster, lead scientist, R&D Biologics for Lonza.
“Biologics, especially complex or multimodal therapeutics, are an area of continued innovation but are difficult to scale up, manufacture, store, and deliver for a variety of reasons,” contends Don Owens, director of product development at TFF Pharmaceuticals.
For example, Owens notes, mRNA vaccines are unstable for prolonged periods at room temperature, are sensitive to shear stress and thermal and oxidative degradation, require ultra-low temperature (-80 ˚C) cold chain storage, and have short expiration dates. In addition, Jens-Christoph Matuszczyk, manager of Process Technology, Cell Culture Technologies Marketing at Sartorius, observes that mRNA in general exists as a variety of constructs, and as it has become popularized, process sub-varieties like modified nucleotides and synthetic templates have arisen to further complicate the space. “Add to this a quickly evolving intellectual property landscape, and the potential paths to pursue for process development become numerous,” he says.
The cost of goods for running mRNA process development experiments is also a challenge, and the existence of many different analytical strategies, each with its own set of benefits and drawbacks, further contributes to the difficulty in achieving a field consensus, Matuszczyk notes. In addition, process automation technology is still nascent, although he does comment that new toolkits are emerging.
Immune cells, meanwhile, suffer from large donor-to-donor variability, making both process characterization and development and the establishment of process control strategies difficult, particularly for autologous applications, according to Matuszczyk. “Process development work is often done using health donor cells, which characteristically could be very different compared to patient cells,” he says.
Lack of automation and real-time analysis is also an issue for stem cell process development and control. This shortage is exacerbated by long lead times for critical raw materials and the fact that process development must be performed for many different steps from media prep to cryopreservation of final product. “Some of the process steps are manual and ‘open’ to reduce cost, and thus technology transfer and scale up can be problematic,” Matuszczyk remarks.
Induced pluripotent stem cells (iPSCs) are another modality on the horizon with process challenges particularly associated with inducement, directed differentiation, and banking steps, comments Avi Nandi, VP and head of process development at the Center for Breakthrough Medicines.
Process development and control of gene therapy products such as recombinant adeno-associated virus (rAAV) pose significant challenges due to their relatively high dose requirements and generally low-yield processes, says Ranga Godavarti, vice president, bioprocess R&D and drug substance supply at Pfizer. Developing processes and controls for small-scale and complex cell and gene (advanced therapies), meanwhile, is challenging due to immature process understanding, especially for early clinical batches, the complexity of gene editing and modification, and the lack of ability to measure critical quality attributes in a timely manner, according to Nandi.
Making accurate measurements is another difficulty, according to Meg Ruesch, vice president of analytical research and development at Pfizer. Because the quality attributes and functionality of gene therapies are challenging to measure, analytical resources can become a bottleneck for process development. “More analytical scientists may be required to develop meaningful assays and test samples as compared with well-known product modalities,” she says.
New process techniques and approaches are needed so there is a steep learning curve where scientists need to learn, expand their knowledge and expertise, and diversify their capability skillset, Potmans concludes. “There are also challenges from the operational perspective regarding the need to establish new platforms and build new manufacturing lines adapted specifically to these new modalities,” she remarks.
Diversity creates complexity
Given that each new product modality brings with it a new set of challenges, Potmans emphasizes that the greater the pipeline diversity, the greater the complexity with respect to processes, analytic, manufacturing, facilities, supply chains, and workforce talent. “These complexities need to be identified, understood, and overcome. That includes ensuring that appropriate raw materials, consumables (resins, filters, cell culture media, single-use products, etc.), and sometimes equipment are available to support the development, manufacture, and the testing of these new modalities, which may require modification of existing or development of new solutions,” she explains.
There is also significant diversity within many of the classes of emerging modalities that adds to the complexity, according to Godavarti. He points to AAV vectors, which have multiple serotypes based on specific tissue targeting. “Each of these serotypes has unique molecular features requiring novel process and analytical methods, and thus development approaches that take more time and effort,” he says.
In addition, process controls tend to be more complex than those for traditional biologics for which there is better process understanding. “For the emerging modalities, we do not yet have that same amount of process and analytical knowledge on what molecular attributes are critical to control, leading to more difficult control strategies,” Godavarti observes. Furthermore, he notes that these new modalities often require novel raw materials that are not as readily available as those for established monoclonal-antibody (mAb) products. Finally, the limited stability of these new modalities at ambient temperature impacts not only the ability to provide final convenient dosage forms, but also requires minimizing hold and processing times during production.
The often-accelerated development timelines for new modalities create additional challenges, particularly given the desire to run multiple scenarios during process development to gather data for regulatory filings and scale up, adds Nandi. “Getting sufficient study data early on via high-throughput characterization can help avoid failures and risk later on,” he says. These issues are frequently compounded by a lack of experience with the bespoke processes and technologies, and as a result there is an ongoing focus on provision of training and motivation to increase retention across process development organizations, Nandi adds.
Bispecific antibodies are just one example highlighted by Webster. “These molecules combine two different fragment-antigen binding (Fab) arms capable of binding two different antigens into a single format. Ensuring that the produced bsAbs have the correct product quality attributes (i.e., correct pairing of heavy and light chains) every time presents unique challenges to process development and control that is not seen for traditional biologics,” he points out.
Experiments for immune-cell processes should ideally be performed at the lowest scale possible to take into account the cost and labor for testing, which can make sampling an issue as well. Process intensification is needed for stem-cell manufacturing to realize better costs and reduce the time and floorspace needed.
Many traditional biologics have been manufactured for over two decades. As a result, a vast amount of [standardized] process and manufacturing experience and knowledge have been accumulated, according to Godavarti. “These molecules fit pre-determined process platforms well, requiring little time for development and enabling a simpler transfer of manufacturing technology. For the new emerging therapies, such process and analytical platforms do not exist,” he states.
As a result, emerging modalities often require custom manufacturing and specialty equipment that may be too complicated or expensive to develop in-house, observes Kayla Hannon, manager, process engineering, with TFF Pharmaceuticals. For similar reasons, this equipment is typically not found at contract development and manufacturing organizations (CDMOs). “These complex modalities require expertise from manufacturers familiar with their novel structures and components and experienced with custom equipment,” she says. “Additionally, biologics in and of themselves are a very diverse class of materials, and each one will have its unique analytical characterization techniques that require specialized equipment, as well as trained personnel to operate that equipment and interpret that analytical data,” Hannon notes.
Finding new platforms and new molecule evaluation models is important and challenging, Potmans agrees. She also stresses that molecular diversification within the pharma pipeline reinforces the need for highly specialized talent. The talent pool is, however, already limited compared to the current and future needs of the biopharma industry, and product diversification will only further stress the talent crunch for these highly specialized resources. Facilities, too, may have different requirements, such as dedicated equipment, more rigorous cleaning procedures, and/or higher biosafety levels. “Manufacturers of new modalities must build flexible and modular manufacturing platforms to manage all of the various cell types and modalities,” Potmans comments.
The lack of platform processes is obvious when considering specific modalities. For mRNA, process and analytical standardization is a challenge due to diverse types of mRNA (messenger, self-amplifying, trans-amplifying, circular, or non-replicating) and templates (microbial-based circular plasmids or synthetic linear DNA), according to Matuszczyk. There are also different encapsulation technologies (lipids, nanoparticles, or lipoplexes). “All these factors contribute to changes in process steps and utilized technologies, making each process require development from the ground up,” he notes. “The high development need due to the diversity of RNA sub-varieties and minimal process standardization is combined with a high cost of goods and a lack of mRNA-specific workforce experience to create real challenges in the field,” Matuszczyk concludes.
Cell and gene therapies also encompass a wide range of highly complex product types that are produced using a wide range of manufacturing processes that are customized and often involve manual operations, Nandi comments. Allogeneic cell therapies support multiple patients and are produced in large batches from healthy donor material, while autologous treatments are derived from and returned to the same, single patient.
In vivo and ex vivo gene therapies also involve different manufacturing processes, with the former usually leveraging viral vectors for gene delivery and the latter as autologous products, says Nandi. Production processes for viral vectors vary widely as well and are challenged by many issues ranging from raw material variability to poor transfection yields to incompatibility with typical aseptic manufacturing unit operations, such as sterile filtration and filling/dosing by pumps. “Effort needs to be spent on process development and optimization before the volume requirements for clinical production can be achieved,” he contends.
Process analytical and digital technologies
The reduction of manual tasks, from both a cost and product quality perspective, becomes increasingly important for the development and manufacturing of novel therapies. For both autologous and allogeneic processes, high-throughput solutions are needed to speed up process development and reduce the time it takes to reach the clinic. Doing so, says Matuszczyk, is achieved by further deployment of higher automation, remote diagnostics, and real-time process monitoring and control that significantly reduce manual operations, thus decreasing contamination risks. “Data analytics tools in particular enable process optimization, further reducing the need for human intervention,” he says.
Implementation of process analytical technology (PAT) solutions, however, remains challenging for novel modalities because there is a lack of regulatory guidance around some automation and sensor technologies, such as spectroscopy sensors that measure multiple analytes, and how to validate chemometric models generated for use in good manufacturing practice (GMP) facilities, according to Matuszczyk.
PAT solutions that enable a quality-by-design approach through the measurement of several analytes simultaneously during design-of-experiment (DoE) studies are therefore important, Matuszczyk observes. “Integrated real-time PAT technology from process development though to manufacturing scale, such as integrated Raman spectroscopy, allows models to be built more rapidly and more easily transferred across scales for real-time monitoring and control,” he comments.
Maintaining sterility in the cell culture environment is, according to Matuszczyk, also of particular importance for the manufacturing of autologous cell-based therapies, where a lost batch has detrimental consequences for the patients to be treated. Therefore, the reduction or elimination of manual steps, enabled by the measurement of metabolites, cell viability, or nutrient consumption with in-line, integrated single-use PAT tools, has the ability to reduce the risk of batch contamination, according to Matuszczyk.
There is also a need for the development of real-time analytical solutions for use in the development of bsAb processes, according to Webster. “One of the main challenges to the process development and control of bsAbs production is the need to have adequate PAT to measure not only the same product quality attributes of traditional mAbs (i.e., N-glycan, charge variants, etc.), but additional technologies to measure product quality attributes unique to bsAbs (i.e., correct pairing, functionality). Current techniques involve offline analytical tools that are effective for development but don’t deliver the level of process understanding and control that at-line and in-line PAT tools have delivered in the past,” he remarks.
“Converting the offline analytics for desired product quality attributes to at-line via automated sampling (i.e., mass spectrometry) or in-line via multivariate process monitoring remains a gap that needs to be addressed,” Webster adds. For some bsAb attributes, PAT tools developed for mAbs can be applied, freeing up time to develop new analytical techniques to measure and monitor the product quality attributes unique to bsAbs.
Digital solutions in general within bioprocessing will be critical to overcoming some of the process development and control challenges presented by the molecular diversity in the clinical pipeline. In addition to PAT, simulation tools will allow for the development of more robust processes more quickly, according to Potmans. “The digital ecosystem of integrated process development that will bring future therapies to the market will focus on process modeling, safe and secure cloud-based artificial intelligence (AI), and limited, but focused, empirical testing,” she contends. “Digital solutions will improve how well processes and products are understood, optimized, and controlled, which will in turn dramatically improve speed and quality while significantly lowering risks as therapies progress through developmental phases,” Potmans explains.
Digital tools for in-silico modeling will also enable prediction of scaleup performance and thus augment process knowledge, Godavarti says. Modeling approaches can, in addition, be extended to analytical areas, such as stability studies.
Technologies such as machine learning will help drive insight from high volumes of data, adds Matuszczyk. “Many companies are implementing data-driven approaches from process development up to commercial manufacturing for faster time-to market and risk reduction, leveraging historical data (e.g., cell lines, clones, product yield, growth profiles) for model-driven approaches that help to faster develop and produce novel vaccines and therapies,” he states.
Reinvention of process development and control
Strategic investments are made in developing novel technology and innovation specific to these emerging areas, Godavarti says. He adds that tools and learnings (high throughput-screening, PAT, digital technologies) from other modalities are applied to these emerging areas to capture efficiencies. “Applying molecular biology-based methods such as sequencing and polymerase chain reaction testing in combination with robust in vitro
cell-based activity assays is key to enabling success,” Godavarti observes.
In mRNA development, for example, Matuszczyk points to new trends in process development. Early-stage screening in very small volumes typically occurs with endpoint analysis as the only data point and larger volumes to investigate more complex process designs using fed-batch mode and PAT, such as Raman spectroscopy for real-time measurement of nucleoside triphosphate consumption. For stem cells, he notes there is a push to fully optimize and characterize processes at the R&D stage and then use integrated, automated, and closed end-to-end processes during process development before tech-transfer to manufacturing takes place to avoid rework, save time, and reduce costs.
Use of a rapid freezing process known as thin-film freezing is another example of process development innovation. This approach enables the formulation of biologics as dry powders with improved stability but without loss of structure and therapeutic integrity. TFF Pharmaceuticals is advancing this technique as a platform technology for improving the formulation of complex modalities, including vaccines, small molecules and biologics, as dry powders. According to Hannon, this approach provides key advantages, including access to more convenient routes of administration due to aerosolization properties (i.e., inhalation or intranasal delivery) and elimination of the need for ultra-cold chain storage due to temperature stability.
The use of statistical process control methods, meanwhile, gives operators confidence that processes are adhering to control parameters, according to Nandi. In addition, if a process is trending in an out-of-specification direction, these systems provide notification so action can be taken before a problem arises, often preventing the loss of valuable product.
Developing a stable production system with all components integrated into a stable producing line is another strategy observed more frequently as emerging modalities mature. The Center for Breakthrough Medicines is developing such a system for viral-vector production that is similar to systems used for manufacturing mAbs, and with it can obtain 5–25 times more product. The company has also established a pilot plant that enables GMP production at non-GMP scale, according to Nandi.
Future challenges for all emerging modalities, says Godavarti, include establishing platform technologies and building sufficient process knowledge so that work flows for process characterization and control strategies can be simplified. Other areas of focus, he adds, include identification of ways to enhance stability, yield, and assurance of the availability of GMP-grade novel raw materials.
Potmans highlights the need to increase capacity at CDMOs with expertise in new modalities. Owens adds that outsourcing and other strategic partnerships require significant planning and financial support by the parties involved. Managing supply chains in general will continue to be an issue for all drug products as well, Owens observes.
“There will continue to be a strong need for the development of innovative formulations, analytical characterization, and advanced process control technologies aimed at improving the development of complex modalities to circumvent some of the slowdown in manufacturing, processing, and logistics,” adds Hannon. Streamlining manufacturing processes is, agrees Matuszczyk, critical for future success (e.g., by increasing the use of automation and closed systems for immune-cell-therapy production processes or reducing the steps needed for stem cell production). Furthermore, researchers are currently looking at in-vivo generation of immune cells, which will add, if successful, a complete new level of complexity to treatment of patients.
The continued advancement of products that utilize alternative routes of delivery is needed to improve patient outcomes in certain diseases and with modalities that might have stability and other challenges when administered via traditional routes, Owens observes. “These other routes may use testing and controls that are faster to implement and use than traditional sterile injectable products,” he explains.While multiple challenges certainly need to be addressed with increasing molecular diversity, it is comforting to think that there is a parallel rise in the numbers and types of innovative solutions to match them.
About the author
Cynthia A. Challener, PhD, is contributing editor to BioPharm International.
Volume 35, Number 11
When referring to this article, please cite it as C.A. Challener, “Process Development and Control for New Modalities,” BioPharm International 35 (11) (2022).