[ad_1]
“It still makes my hair stand on end of the neck, thinking about it,” says Peter Richardson, vice president of pharmacology at BenevolentAI, and describes the time when he realized that baricitinib, a drug previously used to treat arthritis rheumatoid. , could be an effective treatment for Covid-19.
The team at BenevolentAI, a UK company that uses machine learning to aid drug discovery, had been searching through its database of all existing approved drugs, looking for one that could be reused to treat the new coronavirus.
The entire process took only three days.
Ad
“Most pharmaceutical companies had been looking for antiviral drugs, but we approached it from the other extreme and looked at what processes used by the virus could be disrupted,” says Richardson.
Protein kinases, enzymes that speed up chemical reactions in the body, seemed to be a promising area. Some of these regulate how substances can enter human cells: They disrupt them, and the virus may not be able to enter cells in the lungs, heart, and kidneys that it has been so prone to invade.
Baricitinib, a medication developed by Eli Lilley and approved in 2018, stood out because it not only inhibited kinases but also prevented cytokine storms, the body’s extreme autoimmune reactions that have caused so many deaths with Covid-19. It is also likely to be compatible with other medications used to treat the disease, such as remdesivir.Richardson and a team of three part-time researchers identified an initial 370 kinase inhibitor, and then narrowed it down to six that appeared to be more likely to work.
The entire process of finding him had taken only three days.
“Validated the use of AI for this type of problem,” says Richardson. “It would have been impossible for the four of us to do it at that speed otherwise. If you accepted 250 people, you still couldn’t do it at that rate because there would be too many competing ideas. You really can’t do it without an organized knowledge graph and the ability to view it. “
After Richardson and his team. wrote to the Lancet, the medical journal, about his discovery, Eli Lilley contacted and began preparations for a clinical trial of the drug, which will begin in the United States this month in collaboration with the National Institute of Allergy and Infectious Diseases (NIAID).
Richardson hopes that the sometimes notoriously slow clinical trials will begin to produce relatively quick results in this case. “The virus is so fast that clinical trials will never last more than two or three weeks, so we will know very quickly whether patients respond or not.”
Speeding up
The Covid-19 crisis is giving AI companies the opportunity to demonstrate whether they can really accelerate the development of new drugs. Even before the crisis, there were promising signs that AI could get new drugs. through the discovery phase five times faster than conventional methods Drug development generally takes a decade from idea to market, with failure rates of over 90% and priced between $ 2 and $ 3 billion.
AI had promised to cut some of the time and cost, but the first AI-assisted drugs were just beginning clinical trials, and the pharmaceutical industry was still in a “wait and see” phase.
The Covid crisis has sped things up a bit.
With only 2,000 approved medications in stock, none of them could work.
Medical AI companies, including Deargen of South Korea and Hong Kong-based Insilico Medicine, are also conducting research in this area, and Paris-based AI company Iktos is working with SRI International, a chemical company synthetic in California to look for new molecules that can fight the disease. The idea is that Iktos’ deep learning model would design a new virtual molecule and SRI would.
Another UK-based AI company that helps search for Covid-19 cures is Exscientia, which has partnered with Diamond Light Source, the UK’s national synchrotron facility, and Calibr, a division of Scripps Research, to examine nearly all known approved and preclinical medications. to use against Covid-19.
Unlike BenevolentAI, looking through existing drugs is not Exscientia’s specialty. The company most often helps pharmaceutical companies find entirely new treatments, but Andrew Hopkins, founder and CEO of Exscientia, felt it was worth starting with what already existed. “If we could find an already approved drug that works, we could quickly implement it,” says Hopkins, meaning that a drug could be available for use in as little as two to six months.
This involves detecting some 15,000 clinically ready components of the drug to see if they could disrupt the virus, targeting three areas: 3CL protease, the NSP12-NSP7-NSP8 RNA polymerase complex (both vital components for viral replication) and the SPIKE of the protein virus, which interacts with the human cell receptor ACE2 to be able to enter human cells. The first findings are expected in two weeks.
“There is a great opportunity for AI to fill the data gap we have on the virus.”
On the other hand, Hopkins points out that there are only around 2,000 approved medications and it could be that none of them work as effectively as necessary.
“We may not have existing compounds that can address this,” he says.
In that case, the research Exscientia is doing now will help lay the foundation for the next phase: the search for entirely new molecules to use. Data from current studies will already give clues to how the virus works, and Exscientia’s models, which have been designed to run on relatively sparse amounts of data, may point in the right direction.
“There is a great opportunity for AI to fill the data gap we have on the virus,” says Hopkins.
