3 Questions for Natalie Dean, PhD ’14
Assistant Professor, Department of Biostatistics, University of Florida
Dr. Natalie Dean specializes in infectious disease surveillance, survey design, clinical trials, and vaccines. As well as consulting on other pandemic-related projects, Dr. Dean is working with the World Health Organization to design strategies and protocols for testing COVID-19 vaccine candidates in large-scale clinical trials.
Why does it take so long to develop and test vaccines? Where are we in that process for COVID-19?
There are a lot of steps, and all the steps are very important. We start off in highly controlled environments, making sure the products are safe, figuring out the dosing, looking at the immune response, looking at safety, and gradually stepping up to bigger and bigger numbers. In the Phase 3 trials that I work on, thousands of participants are involved to test whether the vaccine protects them from disease.
The paradigm for the pandemic has been trying to overlap steps. You’re taking on a lot more financial risk because you’re basically starting the next step before you even know if you’re going to proceed to it. But it saves a lot of time if it does work. Right now there are a mix of vaccines that are in pre-clinical phase and those that are in Phase 1 or Phase 2 trials—they’re looking at the immune response, they’re looking at safety. Some early candidates are starting to move into Phase 3 trials, with more throughout the summer.
How can the clinical trials timeline be accelerated during a pandemic like this?
After I graduated, I got involved in a Phase 3 Ebola vaccine trial in Guinea. Traditional trials struggled because Ebola spread is very unpredictable. Our trial used a more flexible strategy, ring vaccination, that was very targeted and followed the epidemic as it occurred. Mobile trial teams would drive out to affected communities and enroll people there. We were able to establish that the vaccine was highly effective. After that, the WHO formed the R&D Blueprint for Action to Prevent Epidemics. I’ve been involved in the clinical trials working group, focused on how we evaluate vaccines during outbreaks.
We had a paper come out recently in the New England Journal of Medicine about having a flexible trial structure that allows you to add in new sites as the epidemic evolves and moves.
We’ve been working on a strategy for a multi-country COVID-19 vaccine trial that uses this sort of flexible approach. It’s really emphasizing collaboration. If all these different sites are contributing information to the same trial, then you can get big numbers quickly. It’s those types of adaptive strategies that will shave off time.
You’ve been called upon a great deal to speak with the media about COVID-19. Why is the biostatistical perspective so important?
I have experience with survey design, including designing serosurveys. What does it mean when you hear that some percentage of people in an area test positive for antibodies? I can explain the strengths and limitations of these studies, and I’ve written about how to interpret the results. I’ve spent a lot of time explaining to reporters test sensitivity and specificity and concepts that I teach in some of my biostats courses.
I’m also trying to help people interpret messy data and these complicated, very incomplete pictures of what’s going on. We know that there’s a lot of under-reporting of COVID-19 cases. We know that there hasn’t been enough testing. We know that we’re missing people who don’t have any symptoms. Once someone shows up in the hospital, they were infected as long as two weeks ago, so there are all these lags in reporting. I specialize in emerging infectious diseases, so I’m used to viewing this type of data with caution, but others aren’t.