AI in Drug Discovery – Accelerating The Development of New Treatments
Medicines help save lives, treat diseases, and improve a patient’s quality of life. However, finding new, more effective and tolerable treatments for patients is becoming increasingly challenging.
Being part of the pharmaceutical industry, we are facing challenges in sustaining our drug development programs due to difficulties in identifying new targets alongside spiraling research and development (R&D) costs. Remarkable improvements in computational power, coupled with advances in artificial intelligence (AI) could fundamentally change the drug development process, making it faster and more accurate than ever before.
Why current R&D efforts are becoming increasingly difficult to sustain
The discovery of new molecules has led to the development of pioneering therapeutics to treat a broad range of diseases. Until today, highly qualified scientists from a variety of disciplines work on filtering out a suitable active ingredient from an enormous number of compounds. While we still see a high unmet medical need for some diseases, there is a low success rate in the development of innovative drugs to treat them. Drug development is a lengthy, complex, and costly process, with a high degree of uncertainty. Less than one percent of projects are successful. The entire process, from early research to drug approval, can take 12-15 years and costs two billion Euros on average. Even in late stage clinical development, a high number of projects fail.
The potential of AI in drug discovery
Today, AI is mostly used to identify patterns and new insights, to interpret unstructured data, and to predict and solve problems independently. I am convinced that applied to drug development, AI can accelerate the process by handling large amount of data, which would otherwise have to be done manually by scientists. By leapfrogging this time-consuming step, it will enable scientists to derive structured and unstructured data from multiple sources quicker than today.
In early research, AI contributes to the earlier achievement of project milestones and it can accelerate timelines by enabling a more precise identification of suitable drug targets and lead structures. This, in turn, leads to the improved preselection of innovative lead compounds that have a higher chance of advancing to synthesis in the lab and further evaluation in different experiments. Furthermore, AI supports the prediction of outcomes of laboratory experiments based on data and advanced analytics. With regards to clinical development, AI can be deemed valuable through a reduction in the sample size of clinical trials and shorter duration of clinical studies, resulting in lower costs.
How AI is already being used today
We have already started to apply AI in drug discovery; however, there is still untapped potential. But we will need to expand the necessary knowledge and capabilities within our company and the pharmaceutical industry in general. For me, strategic collaborations with AI-driven companies and academic partners hold the promise to establish a robust, AI-based pipeline as part of our portfolio and address new therapeutic areas. Only then we will be able to unlock the full potential of AI in pursuit of new treatments and ultimately be able to provide new solutions to patients.
Follow Joerg Moeller on LinkedIn for more insights on Bayer Pharmaceuticals and more.