By Exploristics CDSO, Kim Hacquoil
The aim of drug development is to get an effective and safe drug to market to help patients. However, the definition of a successful drug development “program” is dependent on more than just regulatory success. Ensuring the program is executed in an efficient way in terms of cost and time is key. Guaranteeing the whole program is appropriately de-risked at the optimal decision points and at right time can also be the difference between success and failure.
Drug development is hard because it’s complex. Each development program will have different constraints and there are multiple ways to get a drug to market. What may be an appropriate approach for one sponsor may not work for another. What may be the most efficient route for one drug may be inefficient for another. What may be vital for one indication may be irrelevant for another.
Every situation is unique
Often, as sponsors, one of the key questions is “what sample size do I need for the study? “. Since the sample size is a key driver of the time and cost of a clinical trial, it’s important to understand a rough and ready sample size approximation for strategic and portfolio planning. Knowing if the sample size is in the 10s, 100s or 1,000s will have a major financial and operational implications for a project.
Companies may well do an initial (rough and ready) sample size assessment to find a totally impractical or unfeasible study size. They may be constrained by budgets, patient populations (e.g., rare diseases) or time. So where do you go from there? Looking at strategic planning and study design with a slightly different lens enables key drivers behind different designs, assumptions, and situations to be highlighted. Every asset is different and therefore every development program should be tailored appropriately. This will ensure that assets can be positioned correctly, taking into account the competitive landscape and differentiation.
While there are often strong incentives to start a study as soon as possible, it pays to invest a little time and money in thinking about a project more holistically at this stage. Engaging with statisticians who can go beyond initial sample size estimates to assess the overall needs of a project given its unique set of constraints will save time, costs and headaches in the longer term. This more interconnected approach helps to identify important study parameters which can drive success, including sample sizes that are both practical and necessary to deliver meaningful results.
Therefore, every solution is unique
By approaching situations more flexibly and working in partnership with the whole project team including statisticians, studies are more likely to be set up for success, where the definition of success is itself a varying concept. This will involve working in an agile way, scoping out the work initially but also delivering priority pieces first, providing insights to guide decision making and then pivoting and adapting accordingly. This approach makes it easier to add value effectively and efficiently to a project.
So, while it’s ok to initially ask for a “simple” sample size calculation, be ready for further discussion to make sure that the unique case is captured and answered in the best way possible. Statisticians have wide experience working on projects where the solution is not always initially clear, where the question may not have been asked before and where the status quo is being challenged. We are keen problem-solvers, able and ready to think outside the box, push boundaries and work smarter to deliver the right solution. Don’t miss the opportunity to draw more deeply on our expertise to really make a difference to the success of a project. Effective sample sizes are just the first step on the road to more successful studies. To keep on track, it pays to let statisticians be your guide.
Read more:
P value pitfalls in clinical trials
Shifting the development paradigm with innovative trial design
Revolutionising clinical trial design with in silico studies using synthetic data
Statistical Consulting Services
Watch more:
I just need a simple sample size
Can pharma learn from development in other industries?