Scientists at the University of Michigan in Ann Arbor have used artificial intelligence to find alternative pharmaceutical building blocks for 12 drugs under investigation to treat COVID-19 — in an attempt to prevent a drug shortage.
Researchers fear that the same issues with U.S. supply chains that caused sanitizing solutions and other shortages at the beginning of the pandemic may repeat itself with the chemicals needed to synthesize COVID-19 therapeutics and vaccines.
“The WHO has started to discuss who will be the first to receive vaccinations for COVID-19, should they become available,” says Tim Cernak, assistant professor of medicinal chemistry and chemistry. “Just to hit that front group – the essential workers, the sick, the elderly – we will need 4.2 billion doses of a vaccine, because the dosing regimen is at least two doses per person, and there will be wastage. For us synthetic chemists, the folks who actually produce the medicine, that number is mind boggling. We’re in a supply chain crisis.”
The researchers, who are medicinal chemists, say the only approved drug for treating COVID-19 is remdesivir. The drug had an availability of just 5,000 when the outbreak started. As of Aug. 25, the U.S. added 30,000-50,000 new cases each day.
Cernak and his lab were approached by chemical supplier MilliporeSigma to devise solutions to the supply issue. The team combed the federal clinical trials database for drugs currently being considered for COVID-19 treatment, then used the artificial intelligence software Synthia to determine new ways to piece the drugs together.
“A drug has a defined structure,” Cernak says. “Let’s say your drug has 12 atoms in it: three nitrogens and nine carbons. There is more than one way to put those three nitrogens and nine carbons in place. There’s a process synthetic chemists use to break the target down to simple commercially available building blocks. You have to know which reactions will work in the forward direction, to stitch the building blocks together, but we tend to think of the breakdown of the complex drug molecule into simple pieces. We call that retrosynthesis.”
The team traced the synthetic sequences of 12 drugs under investigation and then used crowd sourcing to survey the published and patented synthetic routes to build the 12 drugs. The researchers then encoded the routes in the software and asked it to come up with new recipes.
The approach allowed the researchers to navigate around the starting materials that are already in the supply chain of medicinal target compounds. Each search typically returned multiple proposals, which the researchers then narrowed down according to the overall economics of the starting materials and the overall sequence.
“The route we have found for some of these potential therapeutics might be longer than what’s out there, but in most cases, such as umfenovir, we’re actually finding routes that are shorter and starting materials that are cheaper than what’s currently available,” Cernak says. “We found a way to make the therapeutic bromhexine in one step, which we are pretty stoked on.”