This Algorithm Finds Ways to Make Life-Saving Drugs While Avoiding Patents
A team of Polish and South Korean researchers made a computer program that develops new production methods for existing life-saving drugs without infringing on the drug patents of pharmaceutical companies.
The researchers expect that their new method may be used to create even more robust pharmaceutical patents in the future. On the flipside, though, they acknowledge that their work could be used to "[navigate] around patented methodologies" as well.
Pharmaceutical companies spend tens of billions of dollars each year researching methods of producing drugs that maximize the amount produced, guarantee its purity, and reduce costs. These patented methods of producing drugs--known as synthetic pathways-are arguably a pharmaceutical company's most valuable property.
Creating a new synthetic pathway to a desired molecule, or "target," used to be a painstaking task that required hundreds of hours of research from some of the world's brightest chemists. In 2012, however, a software program called Chematica was released that was able to find new and highly efficient synthetic pathways to arbitrary target molecules by drawing on over 250 years of published research on chemical reactions.
Chematica, which was acquired by pharma giant Merck in 2017, not only greatly expedited the process of discovering new synthetic pathways to a drug, but it also allowed pharmaceutical companies to create airtight patents around newly-developed drugs by locking down multiple synthetic pathways that produce the same substance.
Yet as detailed in a paper published Thursday in Chem, these patents on synthetic pathways were not as airtight as they once seemed. A team of researchers demonstrated that Chematica can also be used to discover new synthetic pathways that avoid the patented part of drug synthesis for a number of incredibly valuable pharmaceuticals. The idea is that these methods, too, may be included in a patent.
"When we started this project I was somewhat skeptical that the machine would find any viable synthetic alternatives," Bartosz Grzybowski, the lead developer of Chematica and co-author of the research, said in a statement. "After all these are blockbuster drugs worth gazillions of dollars, and I was sure that the respective companies had covered the patent space so densely that no loopholes remained."
The technique used by Grzybowski and his colleagues is known as retrosynthesis. Whereas normal synthesis involves beginning with a handful of substances and figuring out how to combine them in ways that will lead to a given target molecule, retrosynthesis begins with the target molecule and works backwards to discover new pathways.
To make this happen, the researchers used Chematica's already-existing synthesis software, but blocked off certain patented steps in the synthesis pathways of various drugs. This forced the system to come up with creative solutions to produce the same molecule that avoided patent restrictions on the pathway.
Gryzbowski and his colleagues tested their retrosynthetic program on three major commercial medicines: linezolid, a last-resort antibiotic; sitagliptin, an antidiabetic drug; and panobinostat, a treatment for multiple myeloma. Taken together, these drugs are worth billions of dollars in annual revenue. Sitagliptin alone brought in nearly $4 billion for Merck in 2015.
While drug companies now have a way to create even stronger patents at their disposal, the researchers note that the same tool could be used more subversively. Meanwhile, determined individuals are still skirting the law to make their own drugs.