Emerging Automated Approaches for Cell and Gene Therapy Manufacture

 
 
A sign of the growth of the advanced cell and gene therapy (CGT) industry is the reframing of the key discussions and critical challenges. Over the past decade the key conversations have shifted from being focused on how to attain regulatory approval to the more recent focus on how to realise true commercial success. To achieve this has required a shift in the way we conceptualize the state of the field whereby we now acknowledge this is no longer simply a clinical or scientific endeavour, but requires an industrial or manufacturing mind-set in order to improve product efficiencies whilst maintaining quality and minimizing costs.

Significant breakthroughs in molecular and cell biology have led to the development of personalised advanced therapeutics with a small, but growing, number of gene-modified cell therapies gaining regulatory approval and entering routine clinical practice. Recent examples include the engineered chimeric antigen receptor (CAR) T-cell products which have recently gained both FDA and EMA approval and have demonstrated significant clinical efficacy against non-Hodgkin lymphoma and pediatric B-cell acute lymphoblastic leukemia.

However, the development and manufacture of advanced CGTs faces a number of translational bottlenecks which must be addressed. Today’s manufacturing processes for CGTs are often manual and open processes which are inherently inefficient, difficult to scale, resource and labour intensive, requires skilled personnel, and necessitates significant human intervention which can result in errors, contamination and final product variation. Critically, the complexity and labour-intensity of current CGT production processes makes the manufacture subsequent use for the treatment of large numbers of patients a significant challenge. CAR-T cell therapies, for example, require the relevant material to be isolated from the patient, target cells selected, activated and transduced/transfected in vitro, expanded and the final cell product formulated, with all unit operations performed in a GMP manufacturing facility. Every day in a GMP suite adds significantly to the cost; since this process can take up to 21 days, depending on the quality of the patient’s original starting cellular material, this represents a significant burden on capital and resources. If the number of patients to be treated increases, the effort and the technology investment costs increase too as current technologies are not optimised for higher throughput. If large patient populations are to be provided for, scalable, robust and automated solutions are required.

Moreover, reproducibility of established processes across multiple manufacturing sites or across scales is challenging, because of variations in personnel, environment, apparatus and expertise. As a result, many current CGTs use a ‘centralised’ model for production, with patient samples being shipped from the point of care to a manufacturing centre, and the resulting autologous therapy shipped back to the clinic. This model, however, has complex logistical challenges, high associated costs and several possible points-of-failure in storage, transport and handling, due to the innate biological constraints and temperature-sensitive nature of CGTs.

Additional complications arise from both the innate variation in source material and differences in the processing of source material (apheresis) at the clinical site. Patient cell samples used to manufacture autologous T cell therapies can vary significantly with respect to composition, based on a multitude of factors including underlying disease, age and genome, as well as cell harvesting technology. Such inter-patient variation means that the duration of manufacturing steps are difficult to predict, and any production system must be flexible, without sacrificing quality. This requires highly trained personnel to execute production steps and make critical process decisions, which can be prone to error and is resource inefficient.

The sector is at something of a turning point. We have an extensive number of CGTs in late-stage clinical trials and therapeutic development companies are very much in the throes of looking to solve the aforementioned scalability, reproducibility, consistency and cost challenges. Automation is on the tip of every tongue and is considered to be the saving grace of the field; yet a simple survey of opinions on what automation is, how and when it is implemented and what should be automated highlights the significant progress that is still to be made. Yet progress on this front is tangible and what were once abstract ideas or technology concepts are now coming to fruition with different approaches to automation being developed in parallel.

First generation automation platforms for regenerative medicine and cell-based therapies such as the SelecT and Cellmate systems (Sartorius Stedim) were designed to replicate human operator processes, akin to the automated platforms used in the automotive industry. These systems were developed to automate otherwise manual actions such as tissue culture flask manipulations and cell culture passaging. Although such systems were associated with high capital costs, they demonstrated increased levels of process consistency for a range of therapeutically relevant cell types and illustrated the benefits of automation for improved reproducibility.

More recent automated, commercially-available expansion technologies for the production of CGTs have attempted to combine multiple process steps in one platform. However, no system currently automates the entire process (viral/non-viral delivery, cell processing and expansion, final fill and finish). Such systems also pose a process bottleneck in that they can only accommodate a single patient sample at a time and are beholden to rate limiting steps of the process such as the cell culture, which for CAR-T production can take upwards of 14 days. Moreover such platforms are limited with respect to flexibility in that they cannot integrate technologies from other suppliers or adapt easily to new technologies and new scientific breakthroughs. This has resulted in an increasing focus toward a modular-based automation approach which would include of a series of modular components which implements one or more functional steps in the overall CGT development process. Modules could potentially follow a standardized layout with defined interfaces for fluid management, power and data supply, product transfer and supply of materials between the modules. The focus on this type of automation would be more on the automation of unit operation and, more importantly, the automation of material from one discrete process step to another. The advantage of such an approach is that it would create discrete components that lead to a flexible pipeline that enables interoperability between manufacturers and technologies, with scope for updating and replacing components as the technology evolves further. This is analogous to “plug and play” that helped revolutionise the computer industry by allowing interchangeability of different components from different manufacturers. Successful implementation of such an approach would represent a step-change in automated manufacturing platforms with true flexibility.

The final step toward a truly automated manufacturing system is the development of an ‘intelligent’ or ‘smart’ production platform which lends itself to the ‘Industry 4.0’ manufacturing paradigm, wherein the system is able to continually adapt and respond based on data exchange and cloud computing to ensure an optimal outcome for each patient sample or each production batch. Although still very much a thought idea, such a concept has generated lots of interest in an otherwise risk-averse industry. For this concept to become a reality would require a significant advance in currently analytical technologies, where novel analytical tools would need to be embedded within the production system. This would enable the development of integrated process analytical technology (PAT) system whereby the critical process parameters are both monitored and modulated throughout the process to maintain the critical quality attributes. The significance of such an advance would be the ability to do real-time release testing in addition to improving comparability and reproducibility across sites. Depending on the nature of the data collected, it would also represent a key means to not just improve process and product understanding, but also result in better patient stratification and therapeutic interventions.

Irrespective of how future technology vendors and therapeutic development companies traverse the CGT field, it is clear that automation, in one guise or another, will play a significant role in improving manufacturability and reduction in overall cost of goods. To what extent ‘intelligent manufacture’ or ‘Industry 4.0’ principles become adopted by the industry remains to be seen, however one thing is for sure, CGTs will continue to generate positive clinical outcomes and automated manufacturing approaches will be necessary to ensure commercial success.

Financial & Competing Interests Disclosure

The author has no relevant financial involvement with an organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock options or ownership, expert testimony, grants or patents received or pending, or royalties. No writing assistance was utilized in the production of this manuscript.

Affiliation

Qasim A Rafiq
Advanced Centre for Biochemical Engineering, Department of Biochemical Engineering, University College London, London, WC1E 6BT, United Kingdom
Email: q.rafiq@ucl.ac.uk
Tel.: +44 (0) 203 108 4420

This work is licensed under a Creative Commons Attribution- NonCommercial – NoDerivatives 4.0 International License.