Start-up companies in the UK are turning to artificial intelligence (AI) for analysis of existing drugs and new molecules. The use of AI in this way has the potential to rapidly accelerate the research and development (R&D) timeline, allowing faster COVID-19 drug development. The COVID-19 pandemic may be a tipping point in companies adopting AI-facilitated drug development, says GlobalData, a leading data and analytics company.

Kathryn Whitney, MSc, Director of Thematic Analysis at GlobalData, says: “BenevolentAI used its AI-based drug discovery platform to identify drugs that could disrupt certain viral entry pathways of COVID-19, thus inhibiting the replication process. In a process that took just three days to complete, over 370 candidates were initially identified, 47 of which are approved for use, with six showing the most promise.

“The drugs identified include Eli Lilly’s baricitinib, which is currently marketed in over 65 countries worldwide as Olumiant to treat rheumatoid arthritis (RA). Baricitinib has been shown with high affinity and at a safe therapeutic dose to inhibit AAK1, a known regulator of endocytosis whose disruption may inhibit entry of the virus into cells. Baricitinib also inhibits the cytokine storm associated with COVID-19 lung tissue damage and acute respiratory distress.”

These finding were published by BenevolentAI in The Lancet in February, highlighting baricitinib as a potential treatment for COVID-19 patients. It would be used before the patient develops significant lung damage, as the drug can act as both an anti-viral and anti-inflammatory. In early April, Eli Lilly announced that it had entered an agreement with the National Institute of Allergy and Infectious Diseases (NIAID) to study baricitinib in a Phase III trial. Results for the trial are expected by June 2020.

Whitney continues, “Another company using AI to search for potential COVID-19 treatments is Oxford-based Exscientia. The company announced at the end of March that it had entered into collaboration with Diamond Light Source, the UK’s national synchrotron science facility, and Calibr, a division of Scripps Research, to screen almost every approved and investigational drug to identify molecules that could become viable treatments for COVID-19.

“The research will use Calibr’s collection of 15,000 clinically-ready compounds, including launched drugs and those shown to be safe in humans, as well as pre-clinical candidates. This technology provides an atomic level of detail to understand how anti-viral drugs can work against viruses. Exscientia’s advanced biosensor platforms will be used to rapidly screen the compounds against three viral drug targets. Exscientia expects the first findings of the research to be available shortly, with full findings available within six to eight weeks.”

“The entire process of developing a drug from pre-clinical research to marketing can take 12–18 years and often costs between $2–3bn, with only about 10% of candidates successfully completing clinical trials and gaining regulatory approval. The use of AI has the potential to rapidly accelerate the R&D timeline, making drug development cheaper and faster, and while some companies have started to invest in the technology, most are still in the watching phase.”