OR WAIT null SECS
Scaling cell therapy production workflows requires the establishment of closed, automated steps.
Over the past decade, a large portion of the resources dedicated to scientific research has been focused on solving widespread unmet medical needs. This is particularly true within cancer, where many patients have limited treatment options. One of the most promising treatment approaches, which is equally as complex as cancer, is cell therapy.
Cell therapies are an emerging modality that are accelerating at breakneck speed. In fact, FDA created an advanced therapies “super office” specifically focused on increasing the volume of reviews for the growing pipeline of cell and gene therapies. Since inception, only a handful of cell therapies have received FDA approval, but there are more than 1200 cell and gene therapies active in clinical trials (1).
As an approved treatment option, cell therapies are relatively new. Novartis’ Kymriah (tisagenlecleucel) was the first chimeric antigen T cell (CAR-T) therapy approved and came to market about five years ago in August 2017 (2). Now, there are seven FDA approved CAR-T drugs on the market, and while this is a major milestone in the history of cell therapy, these CAR-T drugs only cover a small percentage of global unmet medical needs.
Although cell therapy is a promising new modality, improvements in cost and efficacy are needed to unlock the treatment’s potential. The manufacturing process for early cell therapies is labor intensive, with multiple touchpoints. This often results in inconsistencies, contamination, and long throughput times, all of which contribute to subsequent safety concerns and failure modes, as well as the incredibly high costs that hold these therapies back from reaching wider patient populations and realizing their full potential.
Common recent strategies to overcome these hurdles in next-generation cell therapies involve the use of scaling up or out and working with alternative cell populations. Not surprisingly, these innovative workflows both introduce new challenges, such as maintaining performance at scale, and can exacerbate the existing ones if the process isn’t thoughtfully designed.
While it is true that most of the cell therapies in the market today were developed through academic manufacturing processes, the intrinsic challenges of cell therapy extend beyond commercial scaling.
The primary challenge is that peoples’ biologic makeup varies from individual to individual. Whereas many other recombinant protein-based therapies are manufactured using well-characterized cell lines derived from CHO cells that are predictable and enable consistent therapy manufacturing and, thus, consistent therapy performance, cell therapies are not created from well-characterized cell lines and the input material is different every time, yet the objective is still to deliver consistent output. When talking about autologous cell therapy (ACT) specifically, the patients’ cells are the input material into the therapy manufacturing process. This means that the quality of the donor cell material directly correlates to the quality of the output material, which is the therapy itself. Older patients and those that have been fighting challenging diseases often present with cells that are “exhausted,” which is to say they show less responsiveness to stimulation and, in turn, tend to generate weaker immune responses.
Efficacy is also a major challenge. In addition to concerns over the quality of the cellular input, the harsh conditions that immune cells can encounter in tumor microenvironments require that the cells be well equipped with “armoring and camouflage” to withstand the obstacles that prevent them from reaching their target, the tumor. When pursuing ACT, utilizing the patient’s limited number of high performing, qualitatively youthful cells to produce a therapy that is ineffective further limits the treatment options of patients who are already exploring final-resort treatments.
Overcoming these inherent challenges requires not only a considerable amount of expensive quality control, but it also requires adopting instrumentation and workflow processes that minimize the opportunity for failures that are costly—both in actual treatment expense as well as patient outcomes—and support efficiencies that address the critical cost and performance challenges.
With this in mind, significant progress has been made over the past five years to create innovations that accommodate closing, automating, and scaling the manufacturing process.
With the cell therapy market poised for rapid growth in the next decade, new technologies and evolutions in cell therapy manufacturing continue to support the commercialization of high-performing therapies and accelerate the development of the next class of cell therapy breakthroughs.
New innovations brought to market lower treatment costs and boost therapeutic performance by closing the cell therapy manufacturing workflow, thus minimizing manual touchpoints that lead to manufacturing failures.
Across every step of the cell therapy workflow—from collection and isolation, to activation, editing, expanding, and transporting—there are manual process or connection points that are susceptible to failure modes due to contamination or inconsistent practices.
Closing the workflow reduces manual interventions, decreasing the likelihood for human error or contamination to be introduced. When looking specifically at addressing this challenge, there are some steps in the cell therapy manufacturing process where technologies to remove human intervention do not exist. However, there are manual steps that are considered addressable, which characterizes process steps that have traditionally been conducted manually and can now be closed thanks to recent innovation. Cell isolation is one example of an addressable step. Cell isolation has been open and time consuming due to the collection and centrifugation steps, which were conducted manually and required significant repetition to produce a single therapy. Gene editing, whether transduction or transfection, is similar in that many steps have often been carried out by hand in the past. Other examples of steps that are still often open include sampling, though in-line monitoring and auto-sampling will solve this, media formulation, and quality control assays, which don’t affect the therapy itself but rather the treatment release timing and the cost in a major way.
Once the workflow is closed, automation can be introduced to further drive efficiency and help reduce costs. For cell isolation, technologies have been introduced that create an automated environment that consistently delivers greater than 86% isolation efficiency of target cells with 96% purity, all without impacting cell viability. In one specific case, high purity levels of >95% were achieved in a pre-clinical manufacturing workflow that initially struggled to reach cell purity levels higher than 40%. This same environment enabled a cell isolation protocol that can process up to 1 L volume (≥10 billion target cells) in approximately 70–100 minutes, which would historically take three to four hours on average (3).
Automation not only enhances efficiencies by reducing the need for highly skilled technicians to do repetitive tasks, but also increases cell therapy manufacturing success rates and reduces failures caused by contaminations introduced by human touchpoints. Examples where automation has been successfully implemented can be seen throughout these workflows. The most prominent example of this in cell therapy is in the expansion phase, specifically via the introduction of bioreactors, which have made it easy to manipulate the culture and eliminate a traditionally error-prone touchpoint, and manual interventions that have been necessary when using flasks, bags, and vessels. Bioreactors also enable physical connectivity between upstream and downstream steps involved in the process via liquid handling tools such as pumps, sensors, valves, and bubble traps. Cell processing, cell isolation, and genetic manipulation have also been plagued by a heavy amount of labor-intensive steps in the past, but more recent innovations have now made it possible to carry these steps out in a closed fashion as well. Other, less-universal examples of these automation-friendly innovations include solutions to automatically remove paramagnetic beads that are used for cell isolation and activation from the system, as well as the tools for auto-sampling and in-process monitoring.
The speed of manufacturing, which is critical for both therapy developers and patients, is also improved through automation. In studies, new instrumentation that automates the bead removal process delivered significant time improvement. In one case, the automated system accelerated the debeading process from two hours to 29 minutes, representing a 76% reduction, without impacting cell recovery (4) (Figure 1).
While these results are promising, it’s important to consider the approach to closing and automating a cell therapy manufacturing process. Unlike the production of mature modalities, such as monoclonal antibodies, the optimal workflow for the development of cell therapies has not yet been established. As therapy developers work toward closing their workflows and automating as many steps as possible, it’s becoming clear there are two distinct approaches. First, a more rigid approach that seeks to provide a walled off, all-in-one, single solution. The aim here is the create speed and scale by having fixed protocols for each step in the process. The second approach is more modular and focuses on delivering flexibility at each step within the workflow. The sentiment in this second approach rests on the notion that flexibility allows therapy developers to use protocols and instrumentation that best suit a specific workflow for a specific therapy and allows for more in-process adjustments to cater to donor variation. While both approaches are viable, given that scale of manufacturing is going up (allogeneic) and going down (autologous), a modular approach appears to best support a wider range of cell therapy manufacturers pursuing various workflows by giving them the ability to adapt and continue advancing their therapies, whereas the more rigid, singular approach limits their overall flexibility and ability to pivot when necessary.
In parallel to closing the cell therapy manufacturing workflow, tactics to simplify the manufacturing process continue to be explored. In 2022, Novartis showed that they could exclude the expansion process and dramatically simplify the process. By transferring the cell therapy drug to the patient for in-vivo cell expansion, the company shortened this process from 10–14 days down to as little as 24 hours (5).
While the need for closing cell therapy workflows and automating the manufacturing process is apparent, the path forward for advancing cell therapy development remains open for innovation. It is not expected that one approach for cell therapy manufacturing will become ubiquitous; there is a place for allogeneic and autologous, but the infrastructure to support both modes of manufacturing will continue to evolve. With allogeneic manufacturing, scaling the cultivation of donor cells will be immensely important to resolve the economic challenges and make this approach viable long term. For autologous, scaling out the manufacturing process, rather than scaling up the size of batches, is the key. This is where the centralized versus decentralized manufacturing infrastructure debate stems from.
Traditionally, centralized manufacturing of just about anything was the most efficient approach. However, when the product being produced is a biologic therapeutic, additional challenges are introduced. Transportation and logistics, including the regulations that come along with it, make an already complex treatment process even more convoluted. Because of this, the prospect of simplifying delivery through decentralized manufacturing is appealing. Technologies exist to be able to build autologous cell therapy manufacturing at a suitable scale at each of the major oncology institutions across the world. However, the technology for quality control processes is still lagging, which is hindering the buildout of decentralized cell therapy infrastructure.
No matter the manufacturing approach, greater connectivity and integration—both physical connection of instruments and digital connection for data transfer—will help support the next generation of cell therapy manufacturing. Connectivity will help mitigate the variation that is inherent in cell therapy manufacturing. While having a standard workflow is essential for efficiency and scaling, developing an extremely rigid cell therapy workflow can lead to failures due to the large donor-to-donor variations. A modular platform, with closed processes and integrated data sharing will support manufacturing optimization on the fly.
The pace of innovation within the cell therapy space has been awe inspiring. While there are more challenges to be overcome, momentum is already being translated into impact, and there are many reasons to remain optimistic about the modality’s potential in the future.
1. Mikulic, M. Number of Cell and Gene Therapy Clinical Trials in the Global Pipeline as of 2022, by Indication. www.statista.com/statistics, Nov. 23, 2022.
2. Novartis, Novartis Receives First Ever FDA Approval for a CAR-T Cell Therapy, Kymriah (CTL019), for Children and Young Adults with B-cell ALL that Is Refractory or Has Relapsed at Least Twice. Press Release, Aug. 30, 2017.
3. Thermo Fisher Scientific, Achieving High Purity with the CTS DynaCellect Magnetic Separation System. Case Study report (2023).
4. Thermo Fisher Scientific, Streamlining Cell Therapy Manufacturing with Rapid and Reliable Bead Removal. Case Study report (2023).
5. Penn Medicine, Penn Researchers Shorten Manufacturing for CAR T Cell Therapy. Press Release, March 29, 2022.
Evan Zynda, PhD, is senior scientist at Thermo Fisher Scientific.
Volume 36, No. 6
When referring to this article, please cite it as Zynda, E. Addressing Cell Therapy Challenges Through a Modular, Closed, and Automated Manufacturing System. BioPharm International 2023, 36 (6), 26–29.