Within the bioprocessing industry it is often remarked that ‘the product is the process’; or rather, any change in the process, no matter how small, will impact the final product. Consequently, process consistency is seen as essential, and the automation of bioprocesses an appealing prospect as the industry matures. In particular, the automation of upstream processing (USP), where cells are first cultured to a sufficient biomass in a controlled bioreactor environment. The benefits of automation are well known and include enhanced process control, improved product yield and quality, increased regulatory compliance, and reduced operator labour, time, and costs, amongst others (Alford, 2006; Brindley et al., 2013; Challener, 2018; Markarian, 2022).

Automation in USP is not new; indeed, controlling the most critical parameters in growth media is already largely automated. For example, the pumping of base and CO₂ into a bioreactor to maintain pH at physiological levels is usually automatic; as is the sparging of oxygen to control dissolved oxygen (DO); or the addition of surfactants in response to foaming. The automation of these tasks is possible due to a combination of integrated controller units, inline probes, and associated control loops (Rathore et al., 2021).

Other tasks open to automation may be specific to the cell culture mode itself; whether batch,* fed-batch,^† or perfusion.^‡ Feeding during a fed-batch process could happen automatically at defined time points, or when a substrate, such as glucose, drops below a set concentration. The media exchange rate in a perfusion process, typically measured as vessel volumes per day (VVD), could be determined by the concentration of a toxic metabolite. For example, the VVD could increase in response to an increase in ammonia, and so minimise any detrimental effect on cells by ensuring adequate media exchange. At CellRev, we are developing strategies to improve current workflows (batch, fed-batch) by automating the control of cell-to-cell mediated microcarrier aggregation with AggreGuard™ as a complementary strategy to glucose/glutamine supplementation.

Automation of USP outside of process control goes one step further. The automatic sampling of media, followed by automated cell counting, is one such example that demonstrates ‘operator-free’ bioprocessing. This is already a reality for miniature bioreactor set ups such as the ambr15 or ambr250 systems, which both have liquid handlers and can be integrated with third-party equipment (Sandner et al., 2019). A pioneering scenario could even see automation of the entire seed train leading up to bioreactor inoculation; though this would require very sophisticated robotics to carry out the cell expansion workflow.

Whilst automation is appealing, it is incorrect to assume that widespread adoption within USP is imminent. The challenges when adopting fully automated bioprocesses – or even partially automated bioprocesses – include upfront costs, increased process complexity, reduced operational flexibility, and the daunting task of having to integrate equipment and hardware from different suppliers (Ball et al., 2018; Gryseels, 2008). The stage of process development may also determine the likelihood of adoption; automation may not be feasible for uncharacterised or unoptimized processes, such as those found in R&D laboratories, where flexible benchtop systems would be more suited. However, automation may be attractive to those working with commercial processes, which would be explicitly defined and routinely operated. Ultimately, individual laboratories must decide whether their bioprocess, or which part of their bioprocess, is suitable for automation and if the benefits outweigh the drawbacks.

We are already exploring ways to automate some processes in CellRev’s continuous cell manufacturing technology. One example being the use of an online biomass probe as part of a control loop to automatically adjust the concentration of our media additive Continuase^TM on our Livit ACE platform. By doing this, we are introducing partial automation with an aim to move towards true continuous production of cells for applications where the cell is the product, such as cell therapies

* media added at the beginning is only used to culture cells

^† media is periodically topped up with feed

^‡ media is continuously exchanged whilst cells are retained

References

Alford, J.S. (2006) ‘Bioprocess control: Advances and challenges’, Computers & Chemical Engineering, 30(10–12), pp 1464–1475

Ball, O., Robinson, S., Bure, K., Brindley, D.A. & McCall, D. (2018) ‘Bioprocessing automation in cell therapy manufacturing: Outcomes of special interest group automation workshop’, Cytotherapy, 20(4), pp 592–599

Brindley, D.A., Wall, I.B. & Bure, K.E. (2013) ‘Automation of Cell Therapy Biomanufacturing’, BioProcess International, 11(3)s, pp 18–25

Challener, C.A. (2018) ‘Evaluating the Rewards vs. the Risks of Automation’, BioPharm International, 31(9), pp 10–16

Gryseels, T. (2008) ‘Considering Cell Culture Automation in Upstream Bioprocess Development’, BioProcess International, pp 12–16

Markarian, J. (2022) ‘Automating Biopharma Manufacturing’, Pharmaceutical Technology, 46(7), pp 30–33

Rathore, A.S., Mishra, S., Nikita, S. & Priyanka, P. (2021) ‘Bioprocess Control: Current Progress and Future Perspectives’, Life, 11(6): 557

Sander, V., Pybus, L.P., McCreath, G. & Glassey, J. (2019) ‘Scale-Down Model Development in ambr systems: An Industrial Perspective’, Biotechnology Journal, 14(4): 1700766