Drug identified by BenevolentAI shown to reduce deaths in patients hospitalised by Covid-19
A therapy identified using artificial intelligence has been shown to reduce deaths in patients hospitalised with Covid-19 by 13 per cent.
Babraham Research Campus-based BenevolentAI first identified baricitinib as a potential treatment for the virus as early as January 2020 and published its hypothesis in The Lancet and in The Lancet Infectious Diseases, sparking global trials.
Now a University of Oxford team has confirmed results from the Randomised Evaluation of Covid-19 Therapy (RECOVERY) trial, featuring more than 4,000 patients.
It showed that the greatest benefit was seen in seriously ill patients, and was additive to that of dexamethasone and tocilizumab, two other anti-inflammatory treatments previously shown to reduce the risk of death in these individuals.
It increased the chances of patients being discharged within 28 days and cut the risk of their condition worsening.
The results add to evidence from nine clinical trials across 12,000 patients in which the use of JAK inhibitors including baricitinib was found to reduce deaths in hospitalised Covid-19 patients by approximately one-fifth.
Dr Anne Phelan, CSO at BenevolentAI, said: “This story started with our novel AI-derived hypothesis, which quickly led to unprecedented global scientific collaboration from public, private and non-profit organisations around the world. Each has played their part in bringing this game-changing discovery to patients.
“Today that story comes full circle, with another life-saving Covid-19 treatment added to the clinical armamentarium, further validating our hypothesis. The significance of this spans beyond Covid-19: it demonstrates that our technology can fundamentally transform the way we understand disease biology and help uncover new treatment approaches for thousands of diseases.”
Baricitinib is an oral JAK inhibitor owned by Eli Lilly and approved for rheumatoid arthritis.
It has been strongly recommended by the World Health Organization (WHO) at its highest evidence level in the BMJ Living Review and authorised for emergency use as a solo treatment for hospitalised patients by the FDA, or in combination with remdesivir.
Eli Lilly’s placebo-controlled trial, COV-BARRIER, showed baricitinib reduced deaths in hospitalised Covid-19 patients by 38 per cent.