This discovery was made by exploring ways to inhibit the cellular processes that the virus uses to infect human cells. The aim was to identify existing medicines that could potentially stop the progression of COVID-19, inhibit the “cytokine storm” and reduce the inflammatory damage associated with this disease. The team hypothesised that baricitinib may also have antiviral properties, based on its predicted ability to inhibit two members of Numb-associated kinase (NAK) family, AAK1 and GAK.
The team believes that by blocking the virus and reducing inflammation in advanced-stage patients can help to save lives, as it is the severe inflammation that causes death meaning lungs fill with fluid and patients can no longer breathe.
In a recent publication in EMBO Molecular Medicine a team of international scientists from BenevolentAI, Eli Lilly, Karolinska and clinicians in Milan validate this artificial intelligence‐assisted prediction. Using liver organoids infected with SARS‐CoV‐2, they confirm dual antiviral and anti‐inflammatory activities. They also show disease improvement in four COVID‐19 patients who were given baricitinib, paving the way for more rigorous placebo‐controlled trials.
In this editorial comment, Michael B. Schultz, Daniel Vera and David Sinclair review these findings and explore the broader, longer-term implications of the AI-powered study on drug discovery: “Here, the marriage of machine learning and rapid clinical trials provides hope for progress, not only in today’s fight against COVID-19, but in the ongoing fight against acute and chronic diseases”.
Access the review in EMBO Molecular Medicine: