Scientists are combining genetic research, patient data and predictive AI technology to find new ways to solve some of medicine’s greatest mysteries. At the new Institute of Molecular and Computational Medicine (IMCM) at the University of Oxford’s Nuffield Department of Medicine, scientists and clinicians are joining forces with researchers from GSK to improve the speed and development of treatments for complex neurological diseases such as Parkinson’s, Alzheimer’s and ALS. This approach brings together biopharma researchers and world-class medical institutions to forge a new path in drug discovery.
A key role for the patient
Central to this approach is a crucial third partner: the patient.
“Technology is allowing us to bring the patient into R&D like never before,” says Tony Wood, Chief Scientific Officer, GSK. “Today we can measure the patient experience at many levels: from wearables that track changes in movement to devices that monitor sleep patterns and now even genomics which looks on a cellular level. All that creates a massive amount of data – GSK generated 20 billion datapoints last year and we have an AI team in-house making sense of that data. And this is all for the purpose of getting ahead of disease, together.”
Another way to think of it is a totally new research model. “Imagine a laboratory for Parkinson's disease with all the pieces finally integrated: one-third of the contribution comes from patients and their families, one third from GSK researchers and another third from Oxford scientists. Patients and their families play a foundational role in the institute and that makes this collaboration and institute unique,” says John Todd, Director of the IMCM and Professor of Precision Medicine at the University of Oxford.
The UK is fertile ground for research
With current medical technologies, the reality is that the complex nature of neurodegenerative diseases is not well understood. That means the search for more effective treatments requires novel ways to identify, describe and interpret the relevance of subtle yet critical differences in a person’s genetic and biological make up. Putting the patient at the forefront of R&D will deepen our understanding of disease sub-types and enable us to target treatments more precisely, increasing the probability of success.
Enter the UK which is known for its robust life sciences ecosystem and longstanding investments in genetics and genomics that serve as an accelerator for research. GSK complements this with a number of genetically informed drug targets and insights generated from collaborations such as 23andMe, UK Biobank – and more to come from a new UK initiative called Our Future Health. Approximately 70% of GSK targets in research are genetically validated, and published scientific research shows that genetically validated targets are at least twice as likely to become medicines.
Tech drives better predictions
Technology is the glue that brings these puzzle pieces together. By using AI models to search for connections across available patient data sets, scientists can unravel layers of information about the building blocks of disease and discover how to combat it early on.
This begins with identifying genetic factors which raise the probability that an individual will develop a disease and then building on those clues to determine whether, when and how their disease will progress.
These early indicators or predictors of disease help to accelerate drug discovery and development, explains Victoria Higgins, GSK’s Senior Director of UK Academic Alliance Management. “We have a wealth of disparate data—from the patients’ medical history to their DNA and biological samples. If we apply machine learning, we can define patterns to help us to distinguish subtle differences between patient groups.”
Researchers looking into those groups can identify chemical or molecular signals that indicate how the disease develops and progresses. These techniques enable scientists to match potential drug targets with those groups of patients most likely to respond to treatment. By understanding disease more deeply, researchers can prioritise the most promising leads. They can establish proof of concept much earlier in the R&D process and reach conclusions in fewer steps, using AI to check data and fill in the gaps.
“It's about applying these new technologies to identify how disease signatures vary between patients so we can select which patient is more likely to respond to treatment — so-called precision medicine, get earlier reads on clinical trials and get that medicine out quicker” says Higgins.
The implications for patient care are significant — particularly in cases of serious neurodegenerative diseases with patients who often wait years for a definitive diagnosis, by which point their condition may have deteriorated.
More than the sum of its parts
The IMCM recently announced its first research projects focused on Alzheimer’s, Parkinson’s and ALS.
“There's never been a time when the ability to bring a range of technologies together with science has been as advanced as it is right now,” says Tony Wood. “We realised that there was an opportunity to bring it all together to be more ambitious, to be more integrated, and give ourselves an opportunity to really make a difference more quickly. So now we have GSK scientists working side by side with University of Oxford scientists, geneticists and computational biologists — different disciplines working together to think differently about science, tech and talent so we can get ahead of disease.”
The pharma and biomedical sectors turn to data
In spring 2022, Longitude, a Financial Times company, surveyed 300 academics, biomedical and pharmaceutical executives in the UK and US to better understand the landscape of research and development in the sector.
