Tips For Making Better Tech Decisions In Biotech Manufacturing
By Avril Vermunt, EQRx
As startup biotechs and gene therapy companies, how do we strike the right balance between new technology ROI and its impact on P&L? In this article, I discuss considerations and tips for adopting single-use technologies and deciding when to go digital.
Adopting Single-use Technologies
Considerations:
- Off-the-shelf vs. customization
- Balancing design and application
The challenge with single-use is that adoption cases are very established with respect to choosing it, compared to stainless steel, especially in a clinical startup, and for gene therapy there’s a lot of adoption of single-use. However, there’s an increasing number of single-use offerings, and with COVID-19, the competitiveness of the market, and the subsequent supply chain challenges, there is added complexity and diversity. So, how do you select the right single-use technology?
At the beginning of my career, a lot of off-the-shelf equipment was used and just taken at face value. There was never a question of whether it met our needs, but the levels of compliance and understanding around single-use materials have significantly improved. I have worked on customization of single-use technology in the past for certain applications where you have very specific, almost novel, requirements in terms of tech. At the time, we were working closely with our vendors, so it was possible to easily align the technology and the applications. This alignment took time for iterations, from design and prototype to final SKU, and required strong attention to detail. Now, there is a lot more alignment in terms of the range of single-use options and the various needs of the industry.
If you are adopting single-use products off-the-shelf and taking your future needs for granted, it may result in a headache. I advocate up-front risk assessments to understand your applications’ requirements and to ensure you’ve outlined the alignment you’re getting from your single-use design.
Cost considerations are also a challenge. It’s critical to establish a clear problem statement and assess whether your single-use design meets that problem statement. It’s common to be overly conservative, especially when closed processing is not well understood, which results in making longer-term decisions around your operating costs that may not be sustainable or efficient. When designing a single-use system, we have simple line drawings that align perfectly, but single-use systems can get very complicated in reality. They can be large and unwieldly, taking several operators to assemble or install, so understanding the 3D reality of what you’re designing or employing is important. I always endeavor to take the 3D design and use pilot operations as a design lab, making sure you have alignment in practice and not just in theory.
Deciding When To Go Digital
Considerations:
- Data management
- Automation
- Talent pool as a competitive edge
One of the biggest challenges I’ve seen is that a tremendous amount of data is generated and, at least in the first half of my career, we only looked at a tiny percentage of that data to gather meaningful justification for decisions, let alone potential insight into future decisions. The good news is that there are so many more options for guided implementation of solutions. I would say, pick one and start! A lot of organizations spend a lot of time evaluating different options, and it’s wasted time when we’re thinking about the speed of our industry and our current reality. Pick one – whether it’s analytics, digitalization of your records, keeping batch records and SOPs – and start with it. You’ll learn a tremendous amount by doing and those learnings can be incorporated into your next steps. “Be willing to try something” is my biggest piece of advice.
At my previous company, we were evaluating new technologies. Solid standbys like Excel were prevalent. Of course, JMP is a longstanding tool in the industry for running statistics, and there are more sophisticated options for complex engineering tasks like modelling unit operation scale-up and conditions. There are so many intuitive cloud-based tools available now, too. When wrangling data, take care to define consistent terminology and create structure from many different sources. But once in place and maintained, more clever approaches like machine learning can lead to trained tools that reveal more insights and understanding.
Another important piece of data is material attributes and history, not just of your single-use technology but of all raw materials and components. I’ve seen a lot of investigations and projects that look at this retrospectively, and it’s a tremendous amount of effort to put together those data sets and start looking for meaningful information in them. There are projects and examples available where that’s done automatically and the data’s telling us information as opposed to us going to look for answers in our data.
Lastly, it is important to consider how automation is serving our needs. Does it streamline our processes or is it bulky and specialized? Many people provoke emotion by claiming automation is replacing jobs, but, really, it addresses the talent crunch and can be a competitive advantage. In terms of talent, it’s important to remember that other industries and companies in our industry are moving forward. Today’s talent pool does not want to work with inefficient, manual processes or at companies that are not taking advantage of all the key insights available to them. If you aren’t keeping up, you could lose out on really good talent.
Leveraging The Talent Around Us For Growth
Considerations:
- Team approaches
- Transactional vs strategic
- Partners to support core competencies
A philosophy that’s carried me through my career is that a high-performing team can do wonders. My aim is always to understand the collective strengths of a team and leverage them to get even more intelligence than the sum of its parts. With that in mind, I’m at a small company where we can neither afford nor have the time to recruit all the expertise we need. So, we have to evaluate what we have in-house and where we feel strongly that we need to maintain a core competency, then what the gaps are and who we can collaborate with to put together a team that will meet the objectives ahead of us.
As an organization, we are forward-looking about the attributes of a partner that will best work with our team. I would advise all my colleagues to think about what they might need in the future and where they might get it. It is much harder to carefully assess these attributes and establish strong relationships when you’re in crisis mode! Luckily, there’s a huge ecosystem around us, so it’s all about networking to find the partner you need.
These tips were excerpted from a case study presented at a recent virtual event, Bioprocessing Strategies for Operational Efficiency, which included three in-depth case studies and two interactive networking sessions. Details of future events can be found here. You can watch Avril’s case study in full and on demand here.
About The Author:
Avril Vermunt is senior director of manufacturing sciences and technology at EQRx. She has over 20 years’ experience in the field, with experience at a CMO, a Big Biotech, a top supplier, a gene therapy company, and a SPAC-backed startup. The common thread is that her roles have always been at the intersection of process development and manufacturing and focused on how to industrialize our bioprocesses to get products out to patients.