On 3rd October, we hosted our inaugural TechBio event with the theme 'When tech meets bio: collaborations in AI drug discovery'. We brought together speakers from various backgrounds, companies small and large to discuss topics that come up when the different challenges and cultures of Tech and Biotech interact to take drug discovery to the next level.
We were fortunate to have the excellent Uzma Choudry, heading up Biotech/TechBio at Octopus Ventures, share her insights into the current and historic investment landscape in this area for our keynote address. Her insightful comparison with the biotech boom of the 2000s helped to highlight a key message: whilst the investment climate is challenging at the moment, there is hope for better times and it is crucial for companies to take measures now to extend their cash runway, as the current cycle is not going to last forever.
Uzma’s keynote was followed by Biotech partner Sarah Lau providing a concise overview of the essential IP considerations needed to formulate an IP strategy in this area, leaving our audience with a clearer understanding of this critical aspect. Next up, Daniel Pavin from Covington, renowned for his transactional and advisory work in this space, delved into the complex issues surrounding data usage. In a world where data is becoming increasingly valuable, his insights provided a roadmap to help navigate the ethical and legal aspects that can come up in this space. Daniel managed to fit what felt like two days’ worth of insights into an excellent 20 minutes, clearly taking the prize for the fastest rate of information transfer of the day.
Our programme was completed by a lively panel discussion chaired by AI partner Alexander Korenberg and featuring a diverse panel of speakers joining Daniel, including Liz Elmhirst (Achilles Therapeutics), Aline Charpentier (Bruntwood SciTech), Jason Rice (GSK) and Anne Poupon (MAbSilico). The panel had representation from various kinds of companies, from multinational giants to start-ups. They shared their thoughts on the challenges and opportunities that lie ahead in this evolving field, including what different players can bring to the table, how to deal with data in collaborations and corresponding IP strategy whilst also looking ahead to share views on future trends. An insightful discussion with the highly engaged audience ensued and carried over well into the following network reception.
In this fast response article, we'll delve into the key themes and highlights of the day.
As patent attorneys, we particularly enjoyed hearing about IP as a common thread running through the event.
We heard a lot about the often-discussed tension between patenting AI innovations and keeping them secret. We note that this does not need to be a binary choice - a combination of the two can work well for many businesses. Anecdotally, it seems that patenting activity in this field is increasing and this may be driven by the need to satisfy traditional biotech investors, who are more familiar with patents than trade secrets. The drive for engineers to publish their research in this field may also be a factor here, as well as the fact that detectability (often cited as a reason not to patent) may not be as difficult as it is perceived, in view of the large number of publications providing insights into the work of different companies. We also heard about publication or disclosure as a possible defensive tool to the TechBio company.
Whatever the decision on this front, there was universal agreement that it is essential for a TechBio company to have a clear and documented IP strategy from the outset that is driven by business strategy, as well as the ability to stay agile and ensure that the IP strategy evolves as the business strategy and/or competitive landscape evolves.
The importance and value of data in the TechBio space was emphasized by many of our speakers. For traditional biotech or pharma companies, this is perhaps a relatively new consideration and it can be daunting knowing where to start in terms of protecting that data and the outputs from use of that data. Jason from GSK introduced the concept of treating data itself as a form of IP (all the while noting that there is no dedicated IP in data but that it is helpful to consider it in a similar manner).
We heard some really useful practical advice from Daniel on assessing and managing legal and regulatory risk in data-driven projects. Daniel explained the benefits of using a systematic analysis framework to help make sure that the right questions are asked at the outset, including: What data is needed for the project and how will it be used? Is any of the data health information or other personal data? What are the sources of the data (for example, directly from individuals, from a healthcare system, from a biobank or from an online database) and what is the context (for example, a clinical trial)? What contractual or other restrictions apply to the source data?
Daniel then gave an overview of economic models in data deals, including challenges with valuation, and some of the complexities in royalty arrangements.
As has historically been the case in the biotech and pharma sector, collaborations are key when it comes to the TechBio sector. Where AI is being used in drug discovery, we may for example have a smaller tech startup providing services to a more established biotech or pharma company in order to help them develop their therapeutic pipeline.
When it comes to collaboration, it is key to ensure that the parameters of the collaboration are clear from the outset: who is bringing what to the party and what does each get out of it? What are the restrictions on data sharing and uses of that data? What about ownership of IP arising from the collaboration? Ensuring an alignment of expectations and that this is clearly set out in collaboration agreements is essential.
We also considered collaborations within companies: getting the right teams of people with the combined expertise to achieve the best outcomes. This may be in terms of teams of scientists, or it could also apply to teams on boards, each bringing their respective expertise and an understanding of the cultures of the biotech and tech sectors and the expectations of their respective investors. And of course, collaborations between legal advisors, be that teams of patent attorneys with tech and biotech expertise, or teams of IP attorneys working for example with data protection lawyers.
TechBio is a relatively new field and represents a coming together of experts from both the tech and biotech/pharma spaces. This is necessarily going to involve a certain amount of learning for everyone involved: getting to grips with the details of each other’s technical fields and the associated IP issues; understanding the expectations and risk appetites of investors (be they biotech, tech or TechBio investors) and developing knowledge of the recruitment market and the factors that might attract AI/ML engineers to or away from a biotech or pharma company. The regulators and the law, neither of which are known for their speed of change, also need to adapt their approaches to this new interdisciplinary field. We also heard about a lack of precedents on data deals, which represents another opportunity for learning.
One key takeaway is that TechBio brings together two sectors with distinct timelines and expectations. While tech companies often operate on a rapid, week-to-week basis, biotech and pharma industries embrace longer timelines, spanning months and even years. Success in this space can hinge on spanning the differences between these two sectors, leveraging the unique strengths that each brings to the table and making sure that the expectations are clearly set out at the outset. Tech companies traditionally strive to “move fast and break things”, a way of operating that has fostered the plethora of innovations we have witnessed both in business models and technology. As Tech meets Bio, companies need to move a little less fast (and most of all, not break things!) when the safety of patients is potentially involved. It will be fascinating to see what models emerge when the constraints and cultures of Tech and Bio continue to merge.
Get in touch
If you have any questions or would like to discuss any of the above topics, please contact Sarah, Alexander or Kathryn.