By Melinda Young
Several recent changes are driving the pragmatic trial trend nationally, including acceptance from regulators and the growth of big data.
Pragmatic clinical trials measure effectiveness of a study drug, device, or intervention on a wider range of people. Unlike clinical trials that exclude people based on health conditions and other criteria, pragmatic trials study a treatment’s effect on a group of people who are more representative of patients in clinical practice and the real world.
“There are a lot of situations where you have two practices and interventions being used, and they’ve never been tested head to head,” says Scott Kim, MD, PhD, senior investigator in the department of bioethics at the National Institutes of Health (NIH). Kim is scheduled to speak about the research risks of pragmatic randomized control trials at the Human Subject Protection: I Will Survive! conference on Sept. 5, 2019, in Covington, KY.
“Given the desire for real-world evidence, people want to do trials that are comparing interventions that are actually in use and are accepted as indicated for those conditions,” Kim adds.
The goal of a pragmatic clinical trial is to conduct it in a way that tracks what happens in a clinic and in a hospital, Kim says.
“You don’t want a highly complicated research apparatus with lots of new elements and eligibility criteria because that won’t serve the aim of these studies,” Kim says. “You want your research integrated, as much as possible, into every day practice.”
The FDA is supportive of these types of trials, says Mitchell Parrish, JD, RAC, CIP, vice president of legal and regulatory affairs at Advarra. “The FDA is supporting pragmatic trials in terms of validating that the information gathered from these trials is valuable and important for the development of therapies and adoption of therapies once they’re approved,” Parrish says.
But the real driver behind pragmatic trials is data. Researchers seek to do more with existing data, accessing data more broadly.
“That’s a trend that will not go backward,” Parrish says. “With pragmatic trials, you have a component of data collection on a broader scale than what was previously seen in getting data from participants on trial records.”
For example, a pragmatic trial might seek a participant’s pharmacy records to determine their drug adherence. With that data, they can tell when a person was prescribed a drug and how long it took to get a refill, he explains.
“Having more access to information helps to paint a better picture of how the product is working for each person,” Parrish says.
The use of more thorough data can raise confidentiality and privacy concerns. IRBs, researchers, and sponsors should be aware of those risks, Parrish notes.
“Sponsors should provide more confidentiality and privacy sections in the protocol to speak to those safeguards,” he says.
From an IRB’s perspective, informed consent in pragmatic trials can be problematic.
“The traditional regulatory structure of informed consent creates an obstacle to making these truly pragmatic,” Kim notes. “The large apparatus [of regulations] is at tension with the pragmatic aims of these trials because if you want to do this trial in a doctor’s office, how do you do this complicated informed consent?”
The other issue is that risks in pragmatic trials typically are not unknown, so researchers might not desire to go through all the risks as they would if the study involved an investigational new drug or device.
“If you go into your doctor’s office and the doctor is going to start you on a different treatment for blood pressure, most doctors will not go through all the possible risks that could happen in taking that drug,” Kim explains. “They’ll say, ‘Here is the drug. It’s been in use for X number of years, and you should look for these side effects. Take it like this, and call if you have any problems.’”
But while some people in the human research protection arena do not believe that is adequate for research informed consent, the others believe these types of studies are low risk.
“There is this strong intuition that if you are comparing two accepted treatments, such that if someone uses A, it’s a reasonable treatment, and if the person uses B, that’s also reasonable, then there isn’t great data to say which is better,” Kim says. “People in the field say, ‘There’s not that much risk, even if you randomize it.’”
IRBs need to remain flexible in their thinking about pragmatic clinical trials, Parrish suggests.
“Pragmatic trials are meant to represent and capture a more representative population of who will be receiving a product in the healthcare setting vs. a standard clinical trial, which is very well defined with exclusion/inclusion criteria and studying a small part of the population,” he adds.
This is a more representative population that might include people considered vulnerable by IRBs. Pragmatic trial participants could include pregnant women, children, people with economic disadvantages, people who cannot read, and others, Parrish says. (See story in this issue on how IRBs can review pragmatic trials.)
“The thought used to be that you can’t test drugs in these vulnerable populations,” he explains. “But people realized this was very narrow-minded; you would just end up testing products in a certain subset of the population. You wouldn’t know whether they would work in other populations because no one ever gathered data about them in other trials.”
Clinicians need drugs for these vulnerable populations, so the FDA has tried to encourage more trials with pediatric populations and enrolling pregnant women, Parrish adds.
Research involving pragmatic clinical trials is an evolving field, Kim says.
“Research ethicists and regulators all are trying to learn and figure out how to do this ethically and efficiently,” he says. “I don’t think I can give a definitive answer as to what to do, but I would say that at this point we should try to not fall back on rules that are perhaps well-intended, but not sufficient.”