A team of York University researchers has uncovered a critical flaw during the drug development process that, if addressed, could make drug discovery faster, more reliable and less expensive - and ultimately save lives.
The findings by a team led by Sergey Krylov, a Distinguished Research Professor of chemistry at York U, reveal that measurement errors in the early stages of drug discovery can lead to further errors - in decisions - as pharmaceutical companies work to develop new drugs.
Sergey Krylov
"People are used to doing things a certain way, and they just keep at it, even when it's not working," Krylov says. "We need to make people aware of these measurement errors and why it's so important to fix them. It's time to stop and think about the damage these mistakes cause and start making changes."
His team is now working to raise awareness of the findings, urging drug developers and academic researchers to change their practices.
"If we can cut down bad decisions in drug development by even half, we'd see twice as many drugs making it to market," Krylov says. "That means saving twice as many lives and making much better use of time and money."
Pharmaceutical companies spend billions of dollars a year on drug-discovery research. The process begins with identifying disease-related proteins and then searching for molecules that can bind to those proteins and change their function. For example, fever-reducing drugs work by lowering the production of chemical signals that cause the body's internal thermostat to raise its temperature during inflammation.
Vast libraries of chemical compounds are screened and measured to see how strongly each one binds to the target protein. The strength of the binding is quantified and molecules with the strongest interactions go to the next stage of development.
Despite advances in technology, this process remains painstakingly slow, costly and prone to errors.
Krylov's team has discovered that these binding measurements are often grossly inaccurate, sometimes off by as much as a factor of 1,000. The molecules identified as having the strongest bindings are particularly susceptible to these large errors.
"These mistakes mean promising drug candidates get rejected too soon," Krylov says. "When the strongest binders are ruled out, it sets off a chain reaction of missed opportunities and expensive delays in research and development."
To address this, Krylov's team developed a practical method to verify whether specific molecules are measured accurately. Their approach allows researchers to flag compounds for reassessment under more rigorous conditions, potentially salvaging overlooked drug candidates.
The team's discovery, made just a year ago, has yet to gain widespread traction in the pharmaceutical industry or the academic community. However, Krylov is optimistic about the potential for change.
In the coming years, the researchers plan to publish the findings of their ongoing research in scientific journals, engage with industry partners and present their work at conferences. They have also created a free online software tool that evaluates the accuracy of binding measurements and can even reassess data for molecules previously discarded, offering a second chance for overlooked candidates.
"This is a long game," Krylov says. "There's no quick fix here. It's going to take years of education and getting the message out. But if we stick with it, we can really change how drugs are discovered and save a lot of lives."
This story was originally featured in YFile, York University's community newsletter.
