Q&A: Giving Image-guided Stroke Trials a Boost

As co-chair of the Stroke Imaging Research group, Max Wintermark, MD, is spearheading an initiative to build a database of imaging facilities’ capabilities to facilitate image-guided stroke clinical trials. Wintermark, chief of neuroradiology at the University of Virginia Medical School in Charlottesville, recently discussed with Cardiovascular Business.

How did you get involved in this project?
I am interested in acute stroke treatment and believe that imaging can help select patients for accurate treatment. I have been involved in several trials that turned out to be negative. I suspect a component of those results was the lack of imaging standardization. I shared this opinion with a number of colleagues in this field and we created this Stroke Imaging Research consortium, STIR. We had our first meeting in 2007.

You are conducting a survey on imaging centers’ capabilities. Why?
If we want to be efficient performing image-guided stroke trials, then we need to have an idea of the resources that are available. Secondly, it is timely to do it. The NIH [National Institutes of Health]—more specifically, the NINDS [National Institute of Neurological Disorders and Stroke]—is [proposing] a stroke trial network. As part of it, there will be some image-guided trials. In preparation for this network, STIR wanted to have a repository of imaging capabilities at sites across the world.

Will this allow researchers to increase the power of studies?
There are several challenges, but one is always recruitment. If you think about the different types of endovascular devices that are available to treat stroke, it is a field that is evolving very quickly. If it takes you five, six, seven years to complete an appropriately powered trial, it is a big problem because those technologies may have changed. When you are done with your trial, the technology that you tested is obsolete.

Roadmap paves the way

The Stroke Imaging Research group's most recent roadmap recommends:
  • Use of standard terminology that complies with National Institute of Neurological Disorders and Stroke Common Data Elements and other refinements
  • For acute ischemic stroke trials, a standardized imaging assessment of revascularization and a modified Treatment In Cerebral Ischemia score
  • A standardized process to assess the acceptability of ischemic core and penumbral imaging methods as a selection tool in acute ischemic stroke trials
  • A data repository with links to clinical metadata
  • Funding from government agencies to support centers to allow them to obtain a standard dataset with a common imaging-based study protocol
  • Making imaging a required data element in some trials
  • An international stroke clinical trial network comprised of multidisciplinary imaging experts who help to centralize information about the clinical and imaging capabilities of centers that potentially could participate in clinical trials

Does it also facilitate subpopulation analyses?
You want to make sure those trials are valid in different populations. You can reach out worldwide to very diverse populations, including minorities and underserved populations.

Who will oversee the database?
The STIR consortium has a steering committee and a website. It is not just a repository of institutions and their capabilities; we also started pulling images from sites and collecting them centrally. The idea is when you start a trial you need to do a power calculation and test your hypothesis. If you have a specific project in mind you go to the website, write a short description and submit it for review. The steering committee will make sure there is no overlap in projects, and if there is we will encourage collaboration between the different investigators. Once your project has been approved, you can get access to the data collected in the repository.

Does STIR allow participants to share images?
Anyone can contribute images to the repository. If they contribute more than a certain number of images, they can become a member of the steering committee. In the future, if anyone wants to conduct a trial that would require a specific imaging capability, then they could contact STIR. We would search our database and say, ‘Institutions A, B, C and D have exactly what you need and here are the names of the contact investigators.’

Who is paying to maintain the database?
We currently are trying to use our personal time and resources to maintain this database as well as the image repository. As the project grows, it is becoming more and more of a challenge to do so. We are considering different avenues that would allow us to finance the maintenance and growth of this system so it can be helpful in the future.

What avenues?
Perhaps for people who would like to access the database, more than a certain number of datasets, there could potentially be a fee. We are being careful to find a [way] to allow the whole system to be self-sustainable but at the same time not present any impediment to future research and collaboration.

How will you update facilities’ information?
We hope that sites regularly update us on their capabilities as they change. We still plan to reach out to people in our database to ask if anything has changed because the technologies evolve quickly.

Editor’s note: STIR is inviting centers to participate through September in an online survey about their imaging capabilities. The group plans to publish the results. Details are available on the STIR website: https://stir.seton.org.