Binding Kinetics in Drug Discovery

How important is understanding target kinetics in drug discovery?

The study of binding kinetics goes back more than a century and is the foundation of pharmacological theory as we know it today. Yet, although the interactions between drug target, endogenous ligands and exogenous potential drugs are highly complex, the kinetics of most approved and clinically effective drugs are markedly similar.

“Over the past decade a series of articles^1-4 suggested there’s a disproportionately large number of slow-dissociating compounds among approved drugs,” explains Professor Mike Waring, Chair of Medicinal Chemistry at Newcastle University, UK. “It’s hard to know for sure that this type of compound is more likely to survive clinical development testing because you don’t really know what the kinetic profile of all the compounds that went into development were, but the number of compounds that seem to be differentiated based on non-equilibrium binding is quite striking.”

Indeed, in an analysis of the binding mechanisms of approved drugs,¹ the most ‘efficient’ drugs, were those that prevent an equilibrium between the target and its natural ligand being achieved, usually because they bind to the target for a long enough time that the natural ligand is degraded. This is called a non-equilibrium mechanism of action, and it was found to be important to the success of many drugs – from aspirin to the blockbuster drug omeprazole (Losec^®).

Yet, despite the fundamental importance of this characteristic to the success of so many drugs, these desirable binding kinetics were a result of serendipity rather than design. The challenge for pharma was how can we understand these principles and start to rationally optimize drugs to have these properties.

How are binding kinetics measured?

Binding kinetics describe how fast a compound binds to its target and how fast it dissociates from it. So, it measures two things – the on-rate and the off-rate. One of the historical challenges in studying binding kinetics has been the difficulty in measuring them.

Piet van der Graaf, Professor of Systems Biology at Leiden University and Vice President and Head of Quantitative Systems Pharmacology (QSP) at Certara recalls: “Measuring the on- and off-rate used to be a painful, lengthy process which is why, certainly until the early 2000s, most pharma companies weren’t really screening for on/off rates because it took too long and was too costly.”

Two things changed: researchers started to apply in silico modeling methodology to analyze binding kinetics data and modern biophysical techniques like surface plasmon resonance (SPR) emerged, which enabled the direct measurement of on- or off-rates without using radioligands. This technology allowed measurement of the binding kinetics in a much better and faster way. 

“All of a sudden, people could generate lots of data on binding kinetics and hence chemists became interested because they could start to design compounds based on the kinetic information that was generated using these new methodologies,” explains van der Graaf.

Indeed, such was the interest in this area, the K4DD consortium was set up through the Innovative Medicines initiative to look at ways to optimize binding kinetics. Waring, who was involved in establishing the project while at AstraZeneca, explains: “It’s a difficult thing to get to the bottom of. We set out to understand how to optimize kinetics rationally, because most slow-dissociating compounds are discovered by chance and I think that’s still a challenge. It’s still not clear how you design slow or rapid dissociation into a molecule.” The programme recently published a review of its contributions to date⁵ which highlights the development of a combination of experimental and computational approaches to study drug-binding kinetics and learn about structure–kinetics relationships.

Is it essential to know the binding kinetics of your drug?

If so many approved drugs have optimal binding kinetics by chance, is it important to have information about binding kinetics before taking drug candidates further into development?

Many working in drug discovery today would argue that consideration of drug residence time is only relevant when considered in conjunction with pharmacokinetic data.^6,7 Waring adds that while a reasonable number of projects would now use SPR as one of the ways to measure compound affinity and therefore characterize the kinetics as a result, whether the kinetic characteristics of a compound are actually used in decision-making is not clear.

“You can’t uncouple kinetics from affinity and one might argue that, other things being equal, projects would tend to favor the more potent compounds and therefore, by default, they’re likely to select the slower dissociating ones. But it’s still important to assess what the kinetics of the receptor are relative to the pharmacokinetics (PK) of the compound. You would expect PK to be dominant in a lot of cases, but you probably need both pieces of information.”

At Certara, says van der Graaf, they strongly advise people to integrate all their binding and PK/PD data, and they build in silico models that can do just that: “I think the interest in binding kinetics has led to an oversimplification of the idea that basically if you have a compound that has a very slow off-rate, you don’t really have to worry about PK. Using in silico models has the advantage of being able to run thousands of virtual ‘clinical trials’ on the computer, that you simply couldn’t run for real.”

This includes, for example, the hypothetical impact of changing the off-rate or the on-rate of a compound and studying how it affects the dosing regimen, as van der Graaf explains: “Could we convert a drug that you have to take four times a day into a drug that you can take once a day, if we change the off-rate of the compound? These are examples that we can use computers for, which you could never do experimentally.”

Using binding kinetics intel to design better drugs

As if to illustrate the powerful intel that binding kinetics analysis can bring, a recent study⁸ reports fascinating insights about binding kinetics of anti-psychotic drugs (APDs) at the dopamine D2 receptor. It had previously been assumed that extrapyramidal side effects of atypical APDs were due to rapid dissociation from the dopamine D2 receptor. However, by measuring the kinetic binding properties of a series of typical and atypical APDs and correlating this with side effects, a new model of binding was generated that shows association rates, not dissociation rates, are responsible for these side effects. This paves the way for strategies to optimize kinetics at the D2 receptor and improve the therapeutic profile of atypical APDs.

As kinetics data becomes easier to obtain, this is an approach that more and more chemists will use, says van der Graaf. “More and more we’ll start to understand the actual chemistry and biology such that we can design on- and off-rates. Then I think we’ll start to look at how that fundamentally changes the pharmacology – not just in terms of potency or dosing frequency but whether we can fundamentally change side effects too”.

Waring predicts that another future application of binding kinetics will be in developing drugs where shorter durations of action are desirable. For example, in the cancer field, short-acting compounds could give more flexibility in modulating the duration of the pharmacodynamic effects making it easier to manage the efficacy versus toxicity profile of compounds. “If you want a short-acting compound, you wouldn’t want a very slow off-rate, as you won’t be able to control the duration of effect as easily by dose scheduling. So, it might be that in certain cases, reasonably rapid dissociation is desirable. In future, perhaps we’ll be able to use binding kinetics to help us better design drugs with different kinetic properties depending on which mechanism is being targeted and what the likely side effects might be.”

References

1. Swinney, D. (2004) Biochemical mechanisms of drug action: what does it take for success? Nat Rev Drug Discov. 3: 801–808 
2. Swinney, D. (2006) Can binding kinetics translate to a clinically differentiated drug? From theory to practice. Letters Drug Design Discov. 3: 569–574
3. Swinney, D. (2008) Applications of binding kinetics to drug discovery. Pharm Med. 22: 23-34 
4. Swinney, D. (2009) The role of binding kinetics in therapeutically useful drug action. Curr Opin Drug Discov Devel. 12: 31-9 
5. Schuetz, DA, et al. (2017) Kinetics for Drug Discovery: an industry-driven effort to target drug residence time. Drug Discov Today 22: 896-911
6. Dahl, G and Akerud, T. (2013) Pharmacokinetics and the drug–target residence time concept. Drug Discov Today. 18: 697-707
7. Folmer, R. (2018) Drug target residence time: a misleading concept. Drug Discov Today. 23: 12-16
8. Sykes, DA et al. (2017) Extrapyramidal side effects of antipsychotics are linked to their association kinetics at dopamine D2 receptors. Nat. Commun. 8: 763. doi: 10.1038/s41467-017-00716-z.