Artificial intelligence and drug discovery: A conversation with Dr Jane Kinghorn

Dr Jane Kinghorn is the Director of UCL’s Translational Research Office and was the lead for the roundtable on Discovering New Medicines. She sat down (virtually) with roundtable coordinator, Audrey Tan, to share her thoughts on the most salient points of the discussion. Read more about the series and the Insights for Public Policy on the AI for People and Planet website.

Great to speak to you today Jane. Can you tell me about the policy engagement projects are you currently working on?

Within the Translational Research Office (TRO), we’re trying to widen repurchasing networks, as there are some serious funding challenges. We want to take the learning from the COVID-19 experience and see how we can apply that to other areas. That’s something that was discussed during the roundtable as well — how we need to figure out how to sort of leverage the momentum from COVID-19 to change things moving forward.

Was there anything that surprised you about the roundtable discussion?

I really liked the fact that there was a greater understanding from hearing the different perspectives. It was great to see the intersection between the commercial realities and academic perspectives. For example, speaking to the skills gap, the roundtable participants from big pharma were saying they aren’t seeing people coming through with the right skills. However, those based in academia countered that by reminding us that universities are serving much broader communities than just big pharma. I think that, sometimes, people in industry get fixated on what they don’t see without the considering the wider context in which they’re situated. In academia, you have to train people for a variety of different skills.

A second really interesting angle of the day was around cost reduction and equitable access. While it would be nice to see a complete change to the drug pricing system, such that the Government backs basic science research and enforces a more equitable pricing regime, the reality is that it’s not that simple. Additionally, AI is just one tool in the toolbox will not be in a position to reduce the costs anytime soon.

That builds on to the next question was, which is about what you learned?

I think it reinforced my thinking about where AI can, and is, making a practical difference, as well as templates for potential routes for impact in the future. One of the challenges that continues for everyone, no matter where they are in the ecosystem, is that the ecosystem continues to get even more complicated. People are trying to simplify things, but in creating new bodies, regulations and policies, there’s yet another structure and set of ethics to engage with. It’s just not clear and we might inadvertently, be putting up more barriers when we’re trying to break them down. In a way, I was glad to learn that people found it as confusing as I did!

Yeah, I think one of the key takeaways was that the ecosystem is just so complicated right now and it’s almost impossible to navigate. Were there any key lessons or key reflections in terms of the public policy element of AI and drug discovery?

The discussions around tangible things that we could be doing was really interesting. The UKRI has three Centres for Doctoral Training (CDT) in AI-enabled Healthcare Systems based at UCL, Imperial College London and the University of Edinburgh. Those of us working with the CDTs could engage with some of the industry sector partners to do address some of the skills gaps the industry partners have raised. Right now, my team is working with the UCL-based CDT to look at building connections with industry for CASE awards.

What’s a CASE award?

It stands for Industrial Cooperative Awards in Science and Technology and provides funding for PhD studentships where businesses take the lead in arranging projects with academic partners. We’ve also discussed other ways to leverage our relationships with industry partners, potentially through inviting them in to give lectures, seminars, or even career talks. Now that we’re all working virtually, this is something that could be offered within and beyond the UCL community and replicated in the other CDTs.

There is an awful lot to think through in terms of reducing price and costs. AI can reduce cycle times to develop drugs and therapies more quickly, and in theory reduce the costs of developing drugs, but we’re not at that stage yet. From a patient perspective, that’s great, as you should get more therapies coming through in theory, but from a Government payer perspective, we’re not going to see that reduced cost, necessarily, and I think that is where we need some further discussions. During the roundtable we began discussing the issue of who benefits from that cost reductions, but we weren’t able to come to any conclusions. You’ve got people focusing on the money and then you’ve got people focusing on the impact and delivery of therapies to patients. It seems like because the money discussions are quite difficult, these are left to finance directors and legal teams and then it just gets messy. AI won’t necessarily lower costs in drug discovery, but it does have the potential to help teams pick the right target and if you get the target right the first time, every pound that you subsequently invest should be going towards a new medicine. However, the challenge currently is that data scientists do not always have access to the datasets they need with which to identify prime targets.

I think a lot of things in science are still down to serendipity. For example, with the AstraZeneca-Oxford vaccine, and they miss-dosed the first 3000 participants where they only received half the standard dose of the vaccine. When they analysed the data, they found that the people that had half the dose had a better response than those who received the full dose.

Really?!

Yes, so it just goes to show that no matter what kind of advanced technology we have, a lot of it is still down to luck.

Finally, how do we build on the engagement from the roundtable?

JK: Firstly is to work more with our AI-Enabled Healthcare CDTs. Secondly, we are looking to incorporate AI approaches and genetic approaches into our own drug discovery. Here at UCL we are making links and strengthening existing ones. There are a number of projects where we have valuable data from our partnerships with our hospitals and we’re running workshops with patients around using their data for commercial aspects. We’re using these workshops to gain insights from patient-participants and linking them into other activities that are ongoing. Now that we have made the connections in the roundtable, we now need to link these connections into ongoing activities.

Thanks so much for taking the time to speak with me.

It’s been it’s been my pleasure, it’s been really interesting.

_

More about the Roundtables

Jointly supported by UCL Public Policy, UCL Grand Challenges, the Business and Innovations Partnerships team and UCLB from UCL Innovation and Enterprise, UCL hosted a series of roundtable discussions on the topic of artificial intelligence (AI) throughout 2020–2021. This series brought together leading voices in policy, industry, third sector and academia with the aim of stimulating dialogue and forging consensus on how to deliver ‘AI for People and Planet.’

--

--

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store