By Andy Mead, DPhil – Director and Head of Head of Drug Abuse and Substance Use Disorders
Over the years, I’ve seen numerous examples where the sponsor has been left disappointed or frustrated at the drug scheduling decision made on their recently approved drug.
Typically, these discussions involve statements such as “…but they clearly didn’t understand the data” or “…the problem is that the regulators don’t understand the science”. Statements of opinion, that often have little factual basis. So why does this happen? Why don’t agencies and sponsors always see the same information in the same way?
There is no doubt that the regulators have a difficult task – to make a call on whether to recommend restrictions on access and distribution of a novel therapeutic in order to mitigate risk of abuse and misuse of the drug outside of its therapeutic setting. This decision is typically informed by a relatively limited data set of controlled non-clinical and clinical studies and perhaps some limited epidemiological data. And all of this takes place in the context of shifting societal perception regarding drug abuse and changing trends in actual drug abuse.
A sponsor can’t control this decision. What they can do, however, is influence this decision by generating the right data in the right way to ensure that the regulators’ have all the information they need in order to increase the chances of getting this decision right.
So, how exactly can the sponsor do that?
The most important step is to start at the very end of the process – a concept popularised by Stephen Covey in ‘The 7 Habits of Highly Effective People’.
Think about the preconceptions that the decision makers (the regulatory agencies) will have. A novel opioid agonist may well be viewed with significant caution given the very public opioid crisis that has swept the US. Similarly, a novel mechanism for ADHD will likely be compared to approved stimulant medications. Preconceptions will exist based on the mechanism of action, the therapeutic area, the nature of adverse events observed in clinical studies and much more.
What data would it take to change those preconceptions? If it were you, what data would it take to change your mind? This is not about influencing inappropriately. At the end of the day, the most important thing for both sponsor and agency is that we end up with the right decision. This is about increasing the probability that we end up with that right decision.
What are the questions the agency will ask of the data? How do you generate data that specifically addresses those questions? Only with this level of forethought can you help the decision maker come to the same conclusion that you have based on the data.
Which leads me to…
It is critical that studies are designed to the highest level of scientific quality.
Within a single study, regardless of the nature of that study, there is a clear objective, with a design specifically tailored to achieving it. Focus on this is critical.
To take one example: it is possible to design a drug self-administration study in many ways. What schedule of reinforcement should I apply? What training drug should I use? Should I include reference comparators? If so, how many, and which ones?
The relevant regulatory guidance provides insight into the framework that we need to work within, but it doesn’t (and shouldn’t) provide the detailed answers to these questions. This is your opportunity to tailor specific studies to achieve specific objectives. Each study then feeds into the broader overall objective in terms of getting the right decision for your drug.
So, you’ve generated the data. What next?
When designing studies, it can be too easy to think that the majority of the job is done once the study plan is agreed and the study is underway.
How often do you think, at the study design stage, how you’re going to present your data? What data is going to be front and foremost? How will it be presented? Why are you presenting it that way?
Again, this needs careful consideration, and alignment with the overall objectives. You know who you want to influence. You now know, relative to likely preconceptions, how they may need convincing. Did you adjust your data visualisation and analysis to align with this? So, what is the data visualisation that is most likely to convince them?
The same is true for analysis. How you choose to analyse the data should be clearly aligned to the story you are looking to tell. There is no single right way to visualise and analyse data. There is however a right way when it comes to aligning with the overall strategy to achieve the top-level objective of telling a compelling story.
Well, put simply, it might be time to flip things on their head when designing your abuse liability package. Instead of starting with the regulatory guidance, and designing a package of studies that simply achieves the requirements of that guidance, start at the end of the process, and remember the following key points:
This article only begins to touch on concepts important in designing an appropriate abuse liability package. We’ll build on these concepts in future articles, but also consider taking the time to watch the webinar that I gave on this topic earlier this year.
If you would like to know more about our abuse liability assessment expertise, we would love to hear from you. You can get in touch by using the contact form below, or email: firstname.lastname@example.org
For more details or information on how we can support your project please complete the contact form.