Racemisation is an underexplored risk within the pharma industry. Referring to what happens when a drug ‘flips’ inside the body, becoming a mirror image of itself, it can cause a drug to become inert or even dangerous. However, it has always been difficult to identify which drug candidates might be most susceptible.
“There’s a reasonable chance those compounds will never become drugs, because if they don’t fail before, they will probably end up failing in clinical trials as a result of racemisation,” says Dr Niek Buurma, a lecturer in physical organic chemistry at Cardiff University.
Go deeper with GlobalData
Buurma’s team, in collaboration with colleagues at Liverpool John Moores University and AstraZeneca, has developed a possible solution. Calling the problem a ‘remarkable blind spot’ and an ‘overlooked issue’, they have created a quantitative risk assessment tool that can estimate a drug’s chances of racemisation.
This mathematical model, described in the Angewandte Chemie journal in January, has clear implications for drug development. It will allow researchers, both in academia and industry, to predict whether their molecule is at risk.
“People can really assess whether a compound is at risk and whether it’s worth continuing the development of such a compound,” says Buurma.
Unintended effects
Many drug compounds exist in two forms which are mirror images of each other. Although these mirror compounds (known as enantiomers) have the same chemical composition, they are not superimposable, nor are they identical in their behaviours. In the worst-case scenario, one enantiomer may produce desirable effects, while the other does quite the opposite.
How well do you really know your competitors?
Access the most comprehensive Company Profiles on the market, powered by GlobalData. Save hours of research. Gain competitive edge.
Thank you!
Your download email will arrive shortly
Not ready to buy yet? Download a free sample
We are confident about the unique quality of our Company Profiles. However, we want you to make the most beneficial decision for your business, so we offer a free sample that you can download by submitting the below form
By GlobalDataThe most notorious example is thalidomide, introduced in 1957 by the German pharma company Chemie Grünenthal. While it was predominantly a sedative drug, it was also used for nausea and gastritis, and was marketed to pregnant women as a remedy for morning sickness.
Unfortunately, the drug had some unintended effects: its S-enantiomer was teratogenic, meaning it led to birth defects. Between 1957 and 1961, around 10,000 infants were exposed to the drug in the womb, many of whom died. The drug was pulled, and the industry was spurred on to develop more structured regulations in this area.
“Since the thalidomide disaster, the FDA [US Food and Drug Administration] will only approve drugs that have one enantiomer and not both, unless there has been a full analysis of all the effects of both the enantiomers,” says Buurma.
Despite this regulation, there is an ongoing issue, namely that many drugs racemise in vivo. In other words, they flip to the opposite enantiomer when exposed to the chemical conditions of the human body.
“Most of the biological targets that these molecules are designed to hit only have one enantiomer,” says Buurma. “This means one of the drug’s enantiomers will produce a good fit and the other one is likely to fit less well. There will be a lower affinity for the intended target, and the molecules will hit all kinds of other unintended targets, leading to side effects.”
Developing the tool
As Buurma explains, the risk assessment tool was the product of years of preliminary research.
“It started off as a result of a discussion with a friend who worked at AstraZeneca,” he recalls. “We were talking about racemisation risk, and we realised it was very much an ignored topic. So AstraZeneca agreed to fund a PhD student to collect kinetic data on racemisation, and two more PhD students also collected data over about a ten-year period.”
Around the same time, Buurma was put in touch with Dr Andrew Leach of Liverpool John Moores University. A computational chemist, Leach was able to come at the issue from a different angle.
“I do a lot of kinetic and mechanistic studies, and Dr Leach does a lot of computational studies,” explains Buurma. “So, along with the data collected by the three PhD students, we were able explore the correlation between the computational and the experimental data, plotting my data against his numbers in one graph.”
Together, they developed their risk assessment model. The tool is extremely straightforward to use: through adding together a few numbers, researchers can predict their racemisation risk within minutes. If that risk is high, they should be able to identify exactly where their problem lies.
“People can identify which of the functional groups in a molecule make the biggest contribution to the risk of racemisation,” says Buurma. “That means, if you have a compound that looks interesting but is at risk, you can look at the changes you could make to the molecular structure that would suppress the risk and hopefully still keep some of the other desirable properties.”
Identifying dead ends
Buurma points out that, while this model has a clear purpose, this may not have been obvious from the outset. For example, many of the compound groups studied had no obvious utility, their importance only emerging later.
“This study shows how fundamental research can create a clear impact,” he says. “Even research that doesn’t necessarily look applied right at the start is still very important in modern-day science. It’s actually putting all that fundamental research together that allowed us to create a real application.”
In the future, Buurma hopes the tool will be used by drug developers to identify unstable compounds at the design stage. This will allow them to save money, ensuring only the most promising candidates move forward to the next stages of testing.
For the time being, his team is continuing their work on this topic, with a view to better understanding outliers and broadening the model’s scope of application. Already, there are indications the tool is proving useful.
“What we really expect to happen is a lot of dead-end routes are going to be cut right at the start,” says Buurma. “That’s going to happen in the drug development pipeline, as well as in academia – we expect that people who spend a lot of time and effort synthesising molecules will quickly stop synthesising the ones with a high racemisation risk. It will allow people to assess whether the routes they start are likely to be a dead end or not.”
