In brief

Follow up research shows that a drug identified by artificial intelligence could potentially benefit patients with COVID-19.

© Shutterstock

Keeping an AI out for new uses of old drugs

24 Jul 2020

Tipped off by artificial intelligence, a research team is testing whether a rheumatoid arthritis drug could be repurposed as a COVID-19 treatment.

As the worldwide cases of COVID-19 soar past the fifteen-million mark, researchers have devised a clever new way to ‘hack’ the drug discovery process, by turning to artificial intelligence (AI) to repurpose existing drugs for COVID-19.

Using a custom AI developed in-house, UK bioinformatics startup BenevolentAI mined and analyzed clinical data for approved drugs capable of inhibiting both inflammatory damage and infectivity associated with SARS‐CoV‐2. Their AI selected baricitinib, a Janus kinase inhibitor, which has been approved for use as an oral, once‐daily treatment for adult rheumatoid arthritis.

Like any good relay, the researchers next invited collaborators, including Yee-Joo Tan, a Principal Investigator at A*STAR’s Institute of Molecular and Cell Biology (IMCB), to validate their AI’s predictions about baricitinib. By providing biochemical and cellular evidence from in vitro models and clinical studies, the international team showed that baricitinib could indeed stem the cytokine storm and viral propagation seen in hospitalized COVID-19 patients.

Stepwise, the researchers first evaluated the in vitro pharmacology of baricitinib to better understand its anti-inflammatory mechanism. When tested on leukocyte subpopulations from a previous randomized trial for rheumatoid arthritis, baricitinib treatment resulted in a statistically significant decline in IL‐6 levels—a predictor of mortality in severe COVID-19 infections.

Next, they extended their findings to see if baricitinib could also exhibit antiviral activity. Baricitinib showed nanomolar affinities for human numb‐associated kinase (NAK) proteins, which are involved in host viral propagation. Inhibition of NAKs with baricitinib led to a 30 to 40 percent reduction in viral load in coronavirus-infected human primary liver spheroids, validating the AI predictions.

In four patients with bilateral COVID‐19 pneumonia, baricitinib treatment was associated with improvements in clinical, radiologic and viral parameters, along with a rapid decline in the levels of inflammatory markers C-reactive protein and IL‐6. Patient symptoms improved and all patients regained normal lung function. Importantly, baricitinib promoted a progressive rise in the titer of neutralizing antibodies in follow-up studies.

Given the urgency of the situation, this study provides hope to those working in drug discovery that AI predictions may help them to repurpose existing drugs for COVID-19. “This study represents rapid repurposing from AI to the laboratory to a potential bedside therapeutic and supports the testing of baricitinib in randomized controlled trials in COVID‐19 patients,” the authors wrote. “Baricitinib is now in clinical trials and if successful, it gives us more options for COVID-19 treatment,” Tan added.

The A*STAR-affiliated researcher contributing to this research is from the Institute of Molecular and Cell Biology (IMCB).

Want to stay up to date with breakthroughs from A*STAR? Follow us on Twitter and LinkedIn!


Stebbing, J., Krishnan, V., de Bono, S., Ottaviani, S., Casalini, G., et al. Mechanism of baricitinib supports artificial intelligence‐predicted testing in COVID‐19 patients. EMBO Molecular Medicine (2020) | article

About the Researcher

Yee Joo Tan obtained a PhD degree from the University of Cambridge in 1997. She joined the A*STAR Institute of Molecular and Cell Biology (IMCB) as a postdoctoral research fellow and is now a Principal Investigator (joint appointment) and group leader of the Monoclonal Antibody Unit. Tan concurrently holds an Associate Professor position at the Department of Microbiology and Immunology, Yong Loo Lin School of Medicine, National University Health System (NUHS), National University of Singapore (NUS).

This article was made for A*STAR Research by Wildtype Media Group