In general, cell-based therapies involve the delivery of living cells to a patient to replace a missing cell type, the introduction of modified cells with altered function, or the provision of necessary factors to treat a disease by addressing the underlying cause. This emerging approach, which may revolutionize the landscape of medicine, represents a new paradigm in human health care. These therapies offer the potential to dramatically improve treatment of daunting medical conditions such as heart failure, arthritis, stroke, spinal cord injury and cancer.
However, in order for cell-based therapies to achieve their full potential, they must be available at commercial scale with the needed quantity to meet market demand, offer the high quality to function properly and meet evolving regulatory requirements, and come at an acceptable cost to satisfy reimbursement policies and generate acceptable return-oninvestment (ROI) for therapeutic developers.
In the cell therapy market, cell-based products require expansion or manipulation in the lab and are segmented into two categories based on the sourcing of the cells: autologous and allogeneic. Autologous cell therapy products are produced from an individual patient’s cells and from a manufacturing perspective; this involves the production of one batch for each patient. Donor cells are used as the source for allogeneic cell therapy. This means that several thousands of patients might be treated from the same manufacturing batch. Due, in part, to the differences of sourcing and scale of manufacturing (one patient vs. large population), autologous and allogeneic cell therapies require different manufacturing methodologies and strategies to enable commercially viable and affordable therapies.
In this paper we discuss two Myths or Misconceptions about Cell Therapy Manufacturing:
- Automation will solve most of the challenges associated with manufacturing of autologous or personalized cell-based therapies.
- Allogeneic manufacturing is very similar to the Pharma manufacturing model, as commercial production is the result of a manufacturing scale-up
Misconception 1: Automation will solve most of the challenges associated with manufacturing of autologous therapies.
The Cost of the Goods Sold (COGS) is seen as one of the major obstacles to autologous commercialization. Many in the field see automation of the current processes as the ultimate and complete solution to bring autologous manufacturing costs under control. Initially, this perception pushed some companies toward robotic solutions. After all, many scaleout processes at commercial scale have been made economical through automation. In some cases, robotic arms to replace human manipulators have been proposed. However, robotic solutions do not fundamentally change the nature of the procedures, as they are performed in a similar way to manual procedures, with no intrinsic improvement. In short, the robotic approach allows mainly a reduction of labor without gaining other efficiencies or process improvements. The cost reduction associated with lower labor is also counterbalanced with the high expense of these robotic solutions. At early clinical stages, considering the low numbers of lots to be produced, this investment is hard to justify. And, at the commercial stage, the effect of such a solution is partial, as its addresses only one factor — the labor component.
Another challenge of autologous cell therapy is that the source of cells varies as the source patient’s age and medical conditions vary. This variance should be controlled by enabling cell culture parameters to be monitored and adjusted. However, the repetitive robotic approach does not offer the needed control. For these reasons, automation alone, which is efficient in other industries, is not the ideal solution for autologous therapies.
The next generation of automated solutions has moved away from the robotic approach towards new manufacturing concepts. One such concept is the integration of multiple process steps as opposed to automating them as separate steps. For example, new automated devices allow the simultaneous washing and concentrating of cells. In these devices, integrated centrifugation or membrane technologies are used to replace the manual hand washing and volume reduction processes. Another example is the combination of magnetic bead cell selection with washing steps. However, these solutions, albeit much superior to robotic replication of discrete steps, are still not able to solve the commercial challenges of autologous manufacturing and its high COGS – of which the cost of labor is only one contributor amongst many other factors.
An additional major factor affecting autologous COGS is cleanroom space needed per patient. This constraint is a very important design factor that automation solutions need to take into account. The cost of investment for construction and operation of cleanrooms is significant. In order for an automated system to provide a significant cost reduction and solve the COGS challenge, its design should be compact so that it reduces the overall footprint per patient. Therefore reducing both; the labor and cleanroom cost components. However, the current market solutions target mainly one or two units of operations; such as washing and concentration, or expansion and volume reduction. This limited approach means that the many units of operations for the whole process are not fully integrated into a closed and compact system. Additionally, while automation can increase the efficiency of each unit of operation (at least from a manpower standpoint); the overall system may be less efficient from a financial perspective due to the high cost of multiple systems required and the additional cleanroom footprint for their installation and utilization. Moreover, the lack of whole process integration and the inability of the separate systems to communicate with each other mean that costly labor supervision and intervention are needed to move the product from one system to another.
While we concentrate on the COGS in this paper, there are other elements that need to be taken into account to define the optimal system for autologous cell therapy. These other elements include the use of biosensors and intelligent software to allow the process to better adapt to the variability in starting material and the use of closed, singleuse systems to contain the integrated process to facilitate traceability and minimize (cross)-contamination (B, Mahdavi, U. Gottschalk, N. Trainor and T. Smith, The Medicine Maker, Issue #0915 October 13, 2015).
Misconception 2: Allogeneic manufacturing is very similar to the Pharma manufacturing model, as the commercial production is a result of a scale-up.
Many in the field believe that allogeneic manufacturing is a preferable route to cell therapy from a business model perspective. The belief is that this method of manufacturing presents very important advantages in terms of economies of scale, the ability to provide necessary commercial quantity per batch and better control of batch quality. In fact, allogeneic “off-the-shelf” therapies based on adherent-cell platforms may require manufactured lot sizes of 100 billion to a trillion cells, depending on a given indication’s market size and dosage.
However, this perception is not fully accurate based on the current method of allogeneic manufacturing. The current commercial method of cell culture expansion for adherent (anchorage-dependent) cell types – such as Mesenchymal Stem Cells (MSC) – involves traditional planner two-dimensional culture (2D cell culture). 2D cell culture traditionally involves 10-layer vessels that are used as a platform in GMP production of allogeneic therapeutic adherent cells, meaning that the quantity produced is linearly proportional to cell culture surface area. This linearity dependence avoids any true cost reductions during the scaleup, contrary to the pharma model for small molecules and biologics, where the scale-up is performed through the use of bioreactors, resulting in true economies of scale (in terms of cost of equipment, construction, labor, quality and testing) as well as better control of the process and the final product. In fact, the 2D cell culture platforms need extensive manual manipulation as they are limited in size in terms of surface area resulting in the use of multiple small vessels in parallel.
Some attempts have been made to increase the number of 2D layers while keeping the same structural and design architecture. However, these approaches do not present significant benefits in term of economies of scale, except possibly a reduction in manipulation during the process, which is offset by the use of more sophisticated robotic automation. Moreover, manual manipulation of different vessels by different operators may result in quality issues.
True scale-up means higher productivity per volume; this is why some attempts have been made to increase the compactness (more surface per given volume by reducing the space between cell culture plates) of 2D cell culture platforms. While these attempts are a move in the right direction, the improvements are limited and in the range of a 2 to 4X increase in compactness. As an example, in the current clinical pipelines, some important allogeneic therapies have a dosage range between 0.1-1 billion cells targeting patient populations of 500 to 1 million. This means that an annual production of between 50 trillion cells (500,000 patients treated at a dosage of 0.1 billion cells) to 1000 trillion cells (1 million patients treated with a dosage of 1 billion cells) would be needed. This will result in the production of up to 1000 batches of 1 trillion cells per year. For each of these batches, one would need approximately 2000 10-layer cell factories (or up to 2000,000 10-layer cell factories on a yearly basis). The use of other 2D cell culture platforms, such as 40-layer or 120-layer Hyperstacks, does not represent very substantial and significant gains or the real-world ability to produce such cell numbers. The conclusion is that the currently available 2D cell culture platforms will not be adequate in meeting commercial market demand, nor are they a good solution from a cost or quality perspective.
So, is there a better solution for allogeneic cell therapy manufacturing? The answer is yes, the solution is the use of suspension cultures of adherent cells on microcarriers, preferably porous microcarriers that have more surface area in a bioreactor, which allows for more process control and higher productivity. This approach represents approximately an 80- fold improvement in cells-per-volume efficiency, and it also allows the operation of one single vessel instead of hundreds of vessels to produce one batch. Notwithstanding, shifting from static 2D to suspension 3D culture is a significant shift that may potentially cause biological changes to the cells. This should be addressed by examining changes to cell biology and understanding whether such changes are relevant to the mechanism of action of the cell product. For companies already in clinical trials, this approach represents a risk that needs to be balanced with the need to use a manufacturing platform that can accommodate commercial needs. Each case must be examined individually, with the assistance of manufacturing experts in the field who are able to analyze the risks and rewards and estimate the market need and manufacturing compatibility. That being said, allogeneic cell therapies that are currently in early stages should consider using 3D cell culture platforms. The use of 3D platforms will reduce the need for drastic process changes down the road and provide a smoother transition from clinical to commercial production as well as capitalize on economies of scale.
Behzad Mahdavi, Ph,D., MBA , VP of Strategic Innovation & Alliances at Lonza Walkersville Inc.,
Dr. Eytan Abraham is the Head of the Cell Therapy R&T at Lonza Walkersville Inc.,
Dr. Uwe Gottschalks, Chief Technology Officer at Lonza Group Inc.,