<img alt="" src="https://secure.lote1otto.com/219869.png" style="display:none;">
Skip to content

Statistical Leaders and the Future of Drug Development

Martin Frenzel

The landscape of drug development has changed dramatically over the last few decades, and effective statistical leaders have never been more important. Last week, Cytel Research Principal Martin Frenzel moderated a panel discussion on “Leading Drug Development into the Next Generation.” Here, he shares his insight into the critical roles statistical leaders play in drug development and suggests actionable steps that leaders can take to advance the future of drug development.

 

Can you provide a few words on the current state of statistical leadership in drug development, and what makes it so vital to the continued improvement of pharmaceutical drug development?

I would say that statistical leadership has never been more important or more prevalent in drug development. If you look at the past 20 years, we have adapted from fixed, very straightforward statistical designs that could only really answer one or two questions well and that were not very robust to incorrect assumptions. Now we have a full range of adaptive and creative statistical designs to implement in drug development that can help us answer many more questions than we previously could answer, and they can help us answer those questions faster than we were previously able to do. These new approaches have completely changed drug development, but they can't be implemented without the leadership of statisticians. That leadership comes through the education that we have to provide to our cross-functional colleagues in the industry, so they understand what their problem is and how we have worked up a solution to it. And just as important, that leadership comes through the advocacy to use these solutions and be willing to do things differently.

Beyond pure design, statistical leadership has evolved from “I'm going to focus on statistics” to “I'm going to focus on drug development.” That means, for example, expanding our understanding of patient needs and designing trials around those patient needs. It means understanding which patients are most likely benefit. That is inherently a data and statistics problem, and we have led the way in designing trials that focus on those patients who are going to benefit the most. And then finally the strategy of drug development. We don't develop drugs in isolation – there's a competitive landscape, there are patient needs, prescriber needs, and payer preferences. All these things come into play. Maximizing the potential of drugs over all of those areas requires high engagement and leadership from statisticians.

 

What are some keys to successful leadership? What would say are your top three?

The first key is being engaged and wanting to grow your own leadership. Sometimes people show up in industry and they just want to stay in their lane – they can be a successful statistician doing that, but not necessarily a leader. There has to be a desire and a commitment to learning to be a leader, which requires a lot of effort. I think #2 is the ability to develop knowledge and appreciation for all of the other disciplines that are part of drug development. Ultimately, if you want to be a leader in drug development, you need to understand not just what's technically important based on your background, but what's technically important to people who are physicians or PK scientists or commercial leaders in the industry, and you can only successfully lead a cross-functional set of people if you understand their objectives, what they care about, and what they're trying to do within their role in drug development. The third thing that that goes along with the first two is a willingness to be uncomfortable and take yourself out of the zone where you know everything well. If you're going to try to drive others and lead others, you need to engage with others in leadership positions and try to understand their goals for your team or organization. You're almost certainly not going to be an expert in all the things needed to push the team or organization forward, but being a leader is not about being an expert. It's about being OK being uncomfortable when you don't know everything, and trusting those other people that are on your team or are your colleagues to educate you and allow you to help lead and help make the best decisions possible.

 

What have we learned from the COVID-19 pandemic in regards to the role of statistical leaders and leadership?

I think the biggest takeaway from the pandemic is that we are an industry that deals with very difficult problems and once we've solved them, we'd like to keep doing things in that way because we created a solution that we know works. COVID-19 taught us that things in the world are going to happen that are going to disrupt our industry, and when the way we’ve always done things stops working, we have two choices: we can cease development during that time, or we can find ways to do things differently. So just to give a very simple example, when a study is run and the database is then cleaned, locked, and analyzed, there’s a big process of people going to sites to monitor and clean data. It's a very time-intensive approach. During the pandemic, it wasn't always feasible to do those things the way we had always done them. We had to come up with creative ways of ensuring the validity of our data and ensuring the data were clean without being able to be on site. And we need to keep utilizing those advances as we go forward because the pandemic showed us that those approaches can be really efficient and they can reduce the burden on sites, increasing the efficiency with which we can execute drug development programs. So I think the biggest thing is all those creative solutions we came up with, we as leaders have to keep pushing them and not allow us to just go back to the way we were comfortable doing things before.

 

Where are there opportunities for statistical leadership to drive drug development forward?

One thing that comes to mind is the idea that there have been many questions previously that we didn't attempt to answer from a statistical point of view, or did not leverage all of the available data out there to try to answer. A good example is the evolution of real-world evidence in the last 10 to 20 years. Previously, the only evidence that was acceptable for submission of a new indication for a medicine was data that was derived from some kind of clinical trial. But we've turned to this plethora of big databases that we have now, which give us access to real-world patient outcomes, to be able to expand indications for patients that maybe wouldn't have been labeled or wouldn’t have had access before. Utilizing these data to make strong scientific conclusions about risk/benefit requires very thoughtful statistical approaches and leadership from statisticians. To add, these data can also be critical to designing faster trials with fewer patients. As an industry, we’ve developed a number of methods to incorporate real-world data with clinical trial data to expedite decision making, we just have to keep pushing these approaches forward.

There are many other examples, such as target selection and other activities that go on in the preclinical space. These were largely directions chosen based on people's intuition and scientific knowledge, but they maybe didn’t have large databases of molecule screens or associations between various biological pathways and disease activity. Deep and thoughtful analysis of these large data sets can help drive decisions about new directions for the development of a molecule, or new targets to pursue. I think that the more we leverage all of the data that we've generated, which can only be done by leaders really pushing for these newer approaches, the faster and farther we can drive development forward as statisticians.

 

What are some actionable steps that leaders can take in the future?

An element that we discussed a lot in the panel was the responsibility of mid-level and high-level statisticians to grow the younger and newer statisticians. Our leadership can only evolve if those that are coming up, that are going to be the next set of leaders, are given everything that they need in order to grow into the leaders that we need them to be. So things like mentorship, expanding training programs, expanding access to things like formal leadership training as opposed to just kind of learning as you go, I think these are some of the steps that are really important for current leaders to focus on.

 

Thank you very much to Martin for sharing his perspectives in this interview!

 

What's new in Perspectives on Enquiry & Evidence:

New call-to-action

New call-to-action

New call-to-action

contact iconSubscribe back to top