Despite more money going towards developing drugs, the success rate of getting new drugs to market continues to fall. Only one in ten drugs tested in clinical trials gets approved for use. Half aren’t effective enough and a further quarter cause unacceptable side effects in late-stage trials, after many years of development. This makes drugs more expensive, costing the NHS more and meaning effective drugs aren’t as readily available.
Research has shown that drugs designed using genetic information are twice as likely to be successful. This highlights an opportunity to systematically prioritise drug ‘targets’ (the molecule in the body they are designed to interact with) using genetic data from large population studies.
What did we do?
Methods, analytical tools and data sources developed in Bristol are the foundation for this work.
Mendelian randomization uses natural variation in our genes to understand the causes of disease, including uncovering the molecular processes that contribute to diseases developing. We used Mendelian randomization to mimic changes in proteins associated with particular diseases to help understand whether a drug targeting that protein could be effective. It is a powerful approach because it not only provides insights into whether the drug might work but also whether it might have undesirable ‘off-target’ consequences – which is often what causes drug trials to fail. Our approach can help to prioritise which drugs should be taken forward in a randomized controlled trial (RCT), which is considered the gold standard of research trials.
Mendelian randomization is much quicker than RCTs. It also enables drug targets to be evaluated before a drug is developed. This significantly reduces the cost of evaluating possible drug targets. We were able to estimate the effects of over 1,000 potential drug targets on 225 disease traits, identifying good evidence for causal effects in 111 target/trait pairs.
Translation into later phase research, clinical practice and patient benefit
In collaboration with GlaxoSmithKline and Biogen we developed and released a public database containing the full results from our systematic drug target prioritisation. This work has been cited widely and the database queried more than 64,000 times. Our approach to drug target prioritisation has now been adopted by pharmaceutical companies to support their own work in this area.