AsianScientist (Dec. 22, 2020) – Using artificial intelligence (AI), researchers from Singapore have found that a combination of the antiviral drugs remdesivir, ritonavir and lopinavir works best against COVID-19. The results of their study were published in Bioengineering and Translational Medicine.
On average, it takes around a decade to take a new drug from bench to market. But with the coronavirus’ rapid spread across the world, frontline healthcare workers and researchers did not have the luxury of time. Instead of developing new drugs, scientists increasingly turned to tried-and-tested treatments for other diseases—a practice known as drug repurposing. Repurposed drugs, however, are unlikely to be effective on their own.
To efficiently search for effective drug combinations against COVID-19, a team from the National University of Singapore (NUS), Shanghai Jiao Tong University and online learning platform Osmosis (Knowledge Diffusion) turned to artificial intelligence (AI). Specifically, they used a platform known as ‘IDentif.AI,’ or Optimizing Infectious Disease Combination Therapy with Artificial Intelligence.
The team investigated 12 potential drug candidates for their ability to inhibit SARS-CoV-2 virus in monkey kidney cells. The set of candidates included known antivirals like remdesivir, malaria drug hydroxychloroquine and antibiotic azithromycin. With traditional drug screening, a similar 12-drug set tested at ten different doses would have generated one trillion possible combinations.
IDentif.AI, however, revealed that only three dose levels were needed for each drug—cutting down the combinations to be tested to a little over 530,000. Given the drastic reduction in the number of tests, the researchers completed the study within two weeks.
Accordingly, their results showed that the Ebola drug remdesivir performed best on its own, while HIV antivirals ritonavir and lopinavir were ineffective against SARS-CoV-2. However, combining the three resulted in an almost total inhibition of infection, with an inhibition efficiency 6.5 times greater than remdesivir alone. In contrast to previous studies, the team also found that hydroxychloroquine and azithromycin were largely ineffective against the virus.
With IDentif.AI’s proven ability to accelerate the search for optimal drug combinations, the team is now looking to run therapies available in Singapore through the platform and generate novel combinations that can be rapidly deployed and administered in the country.
“With IDentif.AI, we will always be ready to rapidly find optimal therapeutic solutions for the next outbreak,” concluded Professor Dean Ho of NUS, co-leader of the study.
The article can be found at: Blasiak et al. (2020) IDentif.AI: Rapidly Optimizing Combination Therapy Design Against Severe Acute Respiratory Syndrome Coronavirus 2 (SARS‐CoV‐2) with Digital Drug Development.
Source: National University of Singapore; Illustration: Shelly Liew/Asian Scientist Magazine.
Disclaimer: This article does not necessarily reflect the views of AsianScientist or its staff.