Automation & Digitisation
Cell Therapy
Gene Therapy

Overcoming Hurdles for an Automated Cell and Gene Therapy Future

Phacilitate
7 February 2022
SHARE NOW
This Event Report evaluates the key obstacles and barriers that need to be considered when approaching industry-wide automation opportunities for cell and gene therapies.

Meeting once more to analyse the ‘what, when, and how’ of automation in the cell & gene therapy space, today’s meeting saw this group discuss comparability, characterisation, regulatory issues, and at what stage organisations should consider and plan for automation.

To begin, it’s important to be clear why automation matters. Is it a throughput issue, an operator-to-operator variability issue, and are the points of automation the same? Being clear on this and expected regulatory issues before becoming tied to an outcome is not only critical, but helps identify early drivers to make the work happen. Before progressing further, organisations should consider these questions carefully.

Defining Drivers: Asking the Right Questions

“There’s no one size fits all; we have to accept that.”

Agreeing that automation is best adopted early, we come to a fundamental issue: every company in the space has a unique profile, perspective, and set of priorities.

Knowing this, our output may be best focused on educating businesses to help them identify their unique drivers for automation instead of attempting to provide a single answer for every company. There’s no one size fits all; we have to accept that. Structuring these considerations as a decision tree may be an appropriate baseline to proceed from.

It’s also important we remain mindful that automation can only be effectively pursued if an organisation already has a deep process understanding; any less risks creating more issues than it is liable to solve. To this end, it’s common to see companies fail to utilise digital tools that can help them achieve this level of understanding on metrics like throughput and cost.

Process Characterisation and Review: How Much can we Automate?

“As a field, we must acknowledge the challenges we’re facing.”

Are some processes functionally impossible to automate, or can we leverage tools to achieve this across the board?

A key issue when considering this is a lack of data points. This makes it hard to predict processes and sections of them. Business models are also limiting sometimes, with small patient numbers presenting a blocker for process automation. A common lack of understanding of what a final state should be also complicates future work; companies must know, for instance, if they require a variable output for a process.

The gap between pre-clinical animal models and translation to humans is a significant hurdle. With traditional approaches proving difficult, the group has discussed the merits of challenging the paradigm of what potency means. Recently, some companies have been constrained by antiquated potency models that could be reappraised and improved. To this end, finding common denominators between a differentiated IPSC and a CAR-T before addressing regulatory factors could prove ideal.

As a field, we must acknowledge the challenges we’re facing. It’s hard to make a well-characterised cell & gene therapy product, particularly due to how variable final results in patients are.

Automation: When is the Right Time?

“Being ahead on analytics automation can be decisive.”

Using the example of the automation of an alpha beta T Cell product by the company of an attendee, members of the group discussed this question. Key factors that contributed to the success of automation efforts, in this case, were a thorough process understanding and a co-development partnership that allowed access to equipment for experimentation. With such equipment being prohibitively expensive to most organisations, this access proved critical.

This led to the discussion of budget and access as common limitations to early automation. It’s common to see organisations begin to address the subject at points such as late stage 2 development, at which point the difficulty of the task is compounded by increasingly defined processes.

Knowing this, segmenting automation into distinct areas has merit. This can allow an organisation to identify which aspect of automation is commercially viable earlier in process development, such as data collection. Being ahead on analytics automation can be decisive. In time, a modular approach to the process in question can be used to build on early wins, helping to make automation more feasible and pragmatic for work on varying cell modalities and for organisations of varying maturity, budget, and capability.

Comparability: Common Hurdles and Blockers

“A premise of measuring that which is indicative of patient risk is vital.”

Regulatory pushback where automation is attempted is common and expected, with many organisations being burned by initial efforts.

An issue with defending changes to regulatory agencies is that processes have so many moving parts that are affected by innovation. If a pivotal process is changed at an important point within it as part of a push for automation, for example, thorough comparability data is a necessity to defend that change. A further challenge is narrowing operation ranges, with comparability studies often failing due to relying on small values. These hurdles are challenging for organisations and can limit the appetite for automation.

In the experience of attendees of the meeting, pursuing a clear agreement on CQAs and a better understanding of the alignment of quality tests for cellular material can be beneficial. When measuring, a premise of measuring that is indicative of patient risk is also vital; anything less is liable to drive up cost with little real benefit to the final therapeutic product.

Early Adoption and Pre-Clinical Studies: What Needs to Change?

“It’s vital to discuss automation plans earlier in process development.”

Are bridging studies truly practical? Is there a way for organisations to be predictive, such as by using early animal studies?

Establishing a clear dialogue with clients, sponsors, or partners is key. Having the capabilities to do early studies is a trade-off and incurs an opportunity cost through a resource constraint. This means it’s vital to discuss automation plans earlier in process development. Oversampling early processes and accruing an abundance of material also has merit, allowing organisations to run comparability studies and more with greater safety and reliability.


This session was moderated by:

This session was hosted in partnership with Sexton Biotechnologies and AmplifyBio on Tuesday 14th September.

Discover more with our Automation Specialist Group in the network.
Join today.