Last month, the AI-driven pharma-tech company, Exscientia announced that the world’s first Alzheimer’s disease (AD) drug candidate designed by artificial intelligence (AI) was entering Phase I clinical trials. This is the result of a collaboration with DSP, a Japanese pharmaceutical company.
The drug candidate is being assessed for antipsychotic effects associated with AD psychosis, as well as improvements in behavioural and psychological symptoms of dementia. AD is a leading cause of death in the UK and affects an estimated 44 million people.
Lu Rahman spoke to CEO, Andrew Hopkins about the company’s work and the opportunities for AI designed drugs.
LR: Explain the process of designing a drug with AI – how is it different to conventional processes and what are the key benefits of this approach?
AH: At Exscientia, we are re-imagining the entire drug discovery process by delivering key advances across data acquisition, Machine Learning and Generative systems to fulfil sophisticated design objectives. We empower our world-leading researchers by using AI to do the heavy lifting, digesting complex biology data and designing new novel molecules aligned to a defined therapeutic profile. Drug discovery is precision engineering at the molecular scale and our focus is always on the quality of our molecules. Making well-balanced molecules that deliver across competing properties means they are less likely to fail during the subsequent drug development phases. Finding faster and smarter ways to discover new and better drugs is what drives Exscientia.
LR: How well received is the use of this technology globally – are some countries more receptive and why?
AH: As with all disruptive technologies it can take time for its appropriate application to be both understood and appreciated. AI is used globally in other industries but in the pharmaceutical space it would appear uptake has been slower.
LR: What are the challenges that AI use in drug design can bring and how are they overcome?
AH: Implementing AI in drug discovery requires a culture shift with the adoption of new approaches, new processes and ways of doing things. One has to design systems so that AI is at the heart of analysis and design in order to extract maximum benefit. So there are technical barriers when building such systems and in certain cases cultural barriers resisting change. The benefits must be made clear – that thoughtful application of AI and machine learning techniques is about empowering researchers to find new and better drugs, faster for patients who are in need.
LR: What are the main medical needs that offer opportunity for drug design with AI at the moment? Is AD one of them and why?
AH: Globally there is a huge need to design and deliver better drugs faster. At Exscientia, we use AI to design molecules to design high-quality drug candidates that will reach a clinical setting quickly, with drug-like characteristics that should smooth their path through clinical trials.
In the specific case of Alzheimer’s disease, we have delivered a small molecule that exhibits high potency as an antagonist for the 5-HT2A receptor and agonist for the 5-HT1A receptor, whilst selectively avoiding similar receptors and unwanted targets, such as the dopamine D2 receptor. On top of all other drug discovery requirements, selective dual activity is a major additional challenge and difficult to achieve with conventional drug discovery methods. Here application of AI was instrumental to design the candidate molecule.
LR: Can you tell us more about the collaboration with the Alzheimer’s Research UK Oxford Drug Discovery Institute and what it hopes to achieve?
AH: Exscientia is in collaboration with the Alzheimer’s Research UK University of Oxford Drug Discovery Institute (ARUK-ODDI) to develop medicines targeting neuroinflammation for the treatment of AD. This exciting new partnership unites Exscientia’s AI-driven molecular design capabilities with the deep therapy area knowledge and technical expertise of the ARUK-ODDI.
The collaboration will focus on a specific neuroinflammatory pathway implicated in the development of AD. Activation of the NLRP3 inflammasome has been shown to have an important role in AD pathogenesis and, while there have been other efforts to develop anti-inflammatory drugs for AD, targeting NLRP3 inflammasome inhibition in the brain is an innovative therapeutic approach.
Exscientia’s proven AI-driven technology will be applied to efficiently generate high-value novel clinical assets targeting this pathway. Feeding into this are chemical starting points that modulate NLRP3 inflammasome formation identified over years of research by the ARUK-ODDI. Coupling Exscientia’s AI-design systems with the ARUK-ODDI’s biology and screening expertise is expected to speed up delivery of distinct candidate molecules for AD.
LR: How did the collaboration with DSP come about?
AH: Exscientia’s research collaboration with Sumitomo Dainippon Pharma resulted from their interest in our technologies and their application to psychiatric disease that we had previously published in Nature. The collaboration brought together the outstanding productivity of Exscientia’s AI technologies and DSP’s work in monoamine GPCR drug discovery. The partnership with DSP has now led to two molecules reaching clinical stage development.
LR: What are the global opportunities for AI designed drugs?
AH: At Exscientia, we believe all drugs will be discovered using artificial intelligence by the end of this decade, such are the advantages. It is an opportunity to make more molecules and better molecules that have a higher chance of success in the clinic. This approach has the power to transform the pharma industry. Our opinion is that the interface between AI, machine learning and pharma coming together is potentially revolutionary and similar to what we saw when molecular biology and industry came together to feed the biotech revolution 40 years ago.