Written by:
Vice President, Discovery Biology, Discovery Sciences, R&D, AstraZeneca
Vice President, and Head of AstraZeneca's Centre for Genomics Research (CGR), Discovery Sciences, R&D, AstraZeneca
Over recent years, the research community has gained considerable knowledge from advances in DNA sequencing combined with novel technologies that are helping us identify gene variants, and consequently, novel drug targets for the chronic diseases we aim to treat, prevent and even one day potentially cure.
Precision medicine – targeting the right medicine to the right patient at the right time – now underpins our approach to drug discovery. But before we can start thinking about new medicines, it is essential that we grasp the underlying complex biological mechanisms that drive the manifestation, evolution and progression of the chronic diseases we’re focusing on.
Read on to find out how we’re digging deep into disease biology and harnessing data networks and innovative technologies not just to uncover novel drug targets with a higher probability of success, but also to characterise the patient subgroups that are most likely to benefit from treatment innovations.
We’re deriving new discoveries from extensive data sets, applying artificial intelligence, machine learning and advanced imaging to expedite our research, and creating biological models that are more reflective of human disease. It’s all part of our ambition to lead the way in a new era of precision medicines for chronic diseases, driving for better patient outcomes and a healthier and more sustainable future for patients and healthcare systems.
Finding new discoveries from a wealth of data
Back in 2016, we set a bold ambition to analyse two million genomes by 2026. It’s one of the most ambitious sequencing programmes ever undertaken, with the aim to uncover new disease insights and expand the therapeutic world that is available to us. Ongoing advances in genome technologies are helping us produce genetic phenotypes of people with chronic diseases based upon their underlying disease mechanisms. This information is key to identifying the most appropriate genetic target for subpopulations of patients.
Our analysis platform is up to 30 times more sustainable than our peers’ platforms helping us explore diseases faster and more effectively than ever before. We’re making rapid progress and have already analysed over five petabytes of data, identifying rare genetic variants, uncovering new targets, and sharing our insights with the research community.
We’re now expanding our research to explore the relationship between DNA, RNA and proteins over the course of disease. By harnessing new technologies we can investigate largely untapped repositories of data from multi-omics with greater precision including:
- Genomics – maps all human genes and their interactions with each other
- Transcriptomics – reveals which genes are turned on or off within cells
- Proteomics – determines the type and quantity of protein expressed in cells
- Metabolomics – provides knowledge of the cellular activities that have taken place based on the metabolites produced
While genomics provides a static view, the other ‘omics’ are dynamic and subject to change under different conditions. They allow us to probe the more complex and transient molecular changes that underpin the course of disease and responses to drug treatment, creating multi-dimensional models of the whole hierarchical system in healthy, diseased and treated states.
This has far-reaching implications for drug discovery, allowing us to spot and validate potential new drug targets, anticipate toxicity and identify biomarkers that can be used to develop diagnostic tests to guide treatment in the real world.
Accelerating the identification of novel disease targets
We’re combining AI and machine learning to produce knowledge graphs which analyse billions of pairwise relationships to show connections between gene targets, expression and disease. Knowledge graphs can reveal unexplored patterns of disease and are considered less biased than most other approaches used in target identification due to the breadth of information they capture. They also continually evolve as new information is added, creating a ‘living map’ of a disease which we can interrogate to identify novel disease targets. A human could never achieve this level of analysis and insight alone, even if working solidly throughout their whole careers!
Since 2019, we have been collaborating with BenevolentAI to apply knowledge graphs across a range of chronic diseases where high levels of unmet patient need exist, including idiopathic pulmonary fibrosis, chronic kidney disease, systemic lupus erythematosus and heart failure. We’ve already identified three novel disease targets and brought them into our pipeline for further exploration.
Creating biological models more reflective of human disease
Once we’ve identified a gene we believe could be responsible for a disease, we check its function within a relevant biological model. Our scientists, alongside our collaborators at BenevolentAI, have used this Functional Genomics approach to identify five new drug targets during 2021.
A key technology we’re using within Functional Genomics is CRISPR/Cas9 gene editing which introduces specific changes into genes – removing, adding to, or changing the DNA code to reflect the same genetic changes that occur in the diseased state. Using CRISPR we can more closely reproduce the real-life environment and test if our drug candidates achieve the desired response.
Before we harnessed this technology, it would have taken months if not years to produce cell lines; today we can produce them in a matter of weeks. This gives us new opportunities to target the gene variants that we suspect are involved in disease.
Harnessing advanced imaging technologies to see chronic diseases more clearly
Advanced imaging technologies are uncovering the cellular processes that define a disease at a molecular level. We’re harnessing these advances and using them in our clinical trials, redefining endpoints to demonstrate disease modification with targeted treatment.
Through our collaboration with the University of Southampton we’re gaining a better understanding of the processes involved in chronic obstructive pulmonary fibrosis (COPD), an inflammatory condition which restricts airflow from the lungs. Our researchers have used advanced imaging to compare two computerised tomography (CT) techniques and assess their relationship with specific physiological measures of small airways disease. The new techniques can detect disease more effectively than traditional scanning methods and as a result are now being incorporated into our clinical trial design to measure the effect of our targeted precision medicines.
Accelerating towards a better future for people living with chronic diseases
Precision medicine will revolutionise the way we understand, classify and treat chronic diseases. It offers more precise diagnosis, earlier identification of susceptibility to illness, and more effective, targeted healthcare interventions for patients.
We have successfully used precision medicine to select top drug candidates to progress through clinical development, and 80% of our BioPharmaceuticals pipeline now features a precision medicine approach. By harnessing novel technologies in our research and development, we hope to make drug discovery more efficient, helping to improve chronic disease outcomes across the world.
