We've updated our Privacy Policy to make it clearer how we use your personal data. We use cookies to provide you with a better experience. You can read our Cookie Policy here.


Using AI To Get Medicines to Patients Faster

A collection of many different pills.
Credit: Myriam Zilles/Unsplash
Listen with
Register for free to listen to this article
Thank you. Listen to this article using the player above.

Want to listen to this article for FREE?

Complete the form below to unlock access to ALL audio articles.

Read time: 4 minutes

The use of artificial intelligence (AI) is rapidly expanding across many industries, not least drug discovery. It is opening a wealth of new possibilities, through accelerating the analysis of novel molecules and enabling new ways to process data.

To find out more about this rapidly advancing field, Technology Networks spoke with Roger Palframan, head of US Research and digital for Early Solutions at UCB. We discuss how AI is now being used in drug discovery, how it can help medicines reach patients faster and what the future applications of AI might look like.


Sarah Whelan (SW): AI has the potential to influence drug development at many different levels. How is AI influencing early drug discovery?

Roger Palframan (RP): We think about this at UCB in a couple of ways. Firstly, the most important is how to better understand human disease and causal pathobiology, i.e., what pathways and molecules are causing the disease, and how to perturb those pathways to have clinical benefit.

AI can help us to find patterns in data and, increasingly, to infer causality. This can help us prioritize from a long list of potential drug target options and help us make better choices about the targets to go for. At the same time, AI can help us to define the patient population we're looking for across many diseases. This is why when we talk about “the right target, the right drug and the right patient”, the right target and the right patient go together. If we can find a more homogenous, molecularly better-defined patient population, that would help us find better causality concerning the target. This is important, as failure of mechanism is the number one reason for failures of therapeutics. We therefore are investing in technologies and partnerships to try and better understand human pathobiology using AI.

The second point concerns the molecule. At UCB we have three main therapeutic modalities: small molecules, biologics (mostly antibodies) and gene therapy. AI is already helping us become more efficient in how we discover targets for these modalities. AI could potentially help us to find molecules that we would not have found using, say, “brute force” high-throughput screening approaches. So, we're using AI today for small molecule discovery to help us make compounds faster, to choose more quickly which compounds to make and to reduce the number of compounds we're making.

SW: You’ve mentioned that AI has the potential to bring medicines to patients faster. Could you elaborate on this?

RP: We know that drug discovery paths are not linear. We rarely start with one molecular target in one patient population. Often there are pivots and turns along the way, where it’s interesting to explore that molecule in new patient populations, even before it’s been used in the first patient population. In the past, that has been done empirically or through trial and error. With AI, we now consider that there may be more data-driven approaches to be able to identify the right population where intervening on a certain pathway can have a big effect on that population with a high unmet need.

Identifying the right patient population, or a different population to what was first envisaged, is an area where we can get medicines to patients faster and move us away from trial and error.

SW: Do you think that AI has potential for methods like drug repurposing?

RP: That’s a similar approach. By better understanding the pathobiology that's driving a disease in populations that may not have been considered previously, I think we will better identify new patient populations for existing therapies in the future, be they in the clinical development stage or even approved for other indications. So that’s certainly of interest, to be more intentional with the next patient population for that therapeutic.


SW: How can institutions remain agile and responsive to both new AI tools and  AI-generated data? What do you think are the key challenges?

RP: It comes in several different ways – one is the data, and how we access the data and make it available for analytics, be that either internal data, partner data or publicly available data. How do we have data that's of the right quality, and is annotated in the right way for us to be able to use intentionally to both train algorithms and then to use in analysis? That's very important. Then there are areas such as data security and privacy, and how we make sure that we are doing this in the right way, particularly in protecting sensitive patient data and being compliant with regulations.


SW: What are your aspirations for AI in drug discovery in the future? Where do you think this could take us?

RP: My vision for the future is that this just becomes the normal way of working. For example, no longer will we talk about digital organization or an AI function in a company – it will be used as seamlessly as we today use things like Microsoft Office. Or even how we are now routinely video conferencing, whereas before, this would have been quite an exotic thing to do – now it's just the day-to-day.

I think that's how AI will be used in drug discovery; it will become part of what we do and very normal. That will come through seamless integration, especially between the wet lab and the dry lab in silico; that's where the real value will be created.

SW: Is there anything else that you’d like to highlight?

RP: This is such a fast-moving field, so the best thing that you can do is to be agile and adaptable. I think having a vision for the future is great, but I'm not entirely sure how we get there. Therefore, as this evolves quickly, how do we set ourselves up for success to be agile and not constrain the power of AI, especially in early research? How do we free the AI to be able to impact discovery? How do we apply the appropriate risk and framework to some areas, such as patient-facing AI where it naturally needs to be more regulated? But also, how do we free it in other areas to be able to maximize its potential? The danger is applying blanket restrictions on everything that will limit innovation.

Additionally, collaboration is key. As it's moving so fast, it’s important to collaborate to understand where it's making a difference but also to share risk and reward.

Dr. Roger Palframan was speaking to Dr. Sarah Whelan, Science Writer for Technology Networks.

About the interviewee:

Roger Palframan is the head of US Research at UCB, where he has led the development of the company’s gene therapy research platform and is responsible for developing UCB’s US research capabilities. He holds a PhD in immunology from Imperial College London and was a Wellcome Trust Postdoctoral Fellow at Harvard Medical School.