When Robert Califf became the new commissioner of the U.S. Food & Drug Administration, he did so with a plan to emphasize the use of real-world evidence (RWE) in Agency decisions.
So, even though the FDA has been evaluating the use of RWE and issued draft guidance a few months ago, “There’s still quite a bit to do,” Everett Crosland, chief commercial officer at Cognito Therapeutics, told BioSpace.
Incorporating electronic source – eSource – data into existing structures is a looming challenge. “There’s been an explosion in terms of the type and quantity of data that’s available, Crosland said. Digital therapeutics, care management platforms and diagnostic guidances are still evolving, and the structures and rigor around them as data sources are still being developed.
“We need to ensure that we get good data in, but this is a fantastic opportunity to develop new clinical trial endpoints because we can now measure things in a new way,” Crosland said. “We’re seeing data we’ve never seen before, such as the ability to see a patient longitudinally, and with a richness we’ve never had.”
New endpoints can expand the value of clinical studies, helping scientists better understand diseases, their etiology and how to treat them. For example, wearable monitoring devices let caregivers assess patients over time in their normal environment rather than just during their hospital stays or at periodic physician appointments. Alternatively, Crosland said, RWE data can help caregivers assess quality of life by measuring activity levels.
“Using RWE to hit endpoints that are unfeasible in a controlled setting is something to consider,” Derk Arts, M.D., Ph.D., CEO of Castor, told BioSpace from his Amsterdam headquarters. “Using RWE, you can increase participation 100-fold. By sampling more broadly and by reporting data continuously rather than at proscribed periods, you can see how interventions translate into real life. I think you would get some different outcomes.”
“I see a lot of experiments about incorporating sensor data into clinical trials,” Arts continued. The data may amplify the trial by accommodating patient-contributed data from Fitbits or Apple watches, for example. Such data points as resting heart rates, number of steps taken per day and stress levels “will become very valuable, but we are uncomfortable incorporating it into a registration trial, so it may be separated out.”
It also could serve as historic, baseline, data for certain interventions. Arts, for example, said he has many years of data from his own wearable that’s not being used. Others do, too.
Several organizations are using RWE to develop digital twins of study participants. The industry “is in the very early adopter stage. Stakeholders want more data validation of digital twins,” Crosland said.
He suggested digital twins could be useful as synthetic control arms, reducing the need for patients to receive placebos rather than an innovator therapy. The goal is to use synthetic arms to inform studies in the real world.
It probably will be a long time before synthetic arms are used in registration trials routinely, Arts predicted. Yet, “BioMarin did a single-arm trial and brought in an RWE cohort matched to the trial population. It was only missing the placebo effects, which can be significant.”
The placebo effect, however, can be modeled. Using digital twins and other large databases, trial sponsors can develop synthetic patients, account for the placebo effect and still gain an understanding of the likely effectiveness of the therapy before it is administered to actual patients. “A lot of people feel placebo treatment is unethical,” Arts pointed out. By using RWE to develop synthetic arms for clinical trials, the need for placebo controls could be reduced.
Looking forward, Crosland and Arts predicted that RWE also will play a role in label expansion. “That’s happening to some extent today and will happen more often because of the ability to measure patients longitudinally and passively,” Crosland said. The insights that emanate from wearables also can identify risks much faster and cost-effectively than traditional methods. He said he expects another FDA announcement regarding RWE in the coming weeks.
Additional goals include using RWE to shorten the drug development process and to also change physicians’ practice behaviors. Incorporating RWE into real-time predictive models using data from multiple sources enables drug developers and physicians get a better notion of whether a particular treatment makes sense for an individual patient.
“The FDA is taking a systematic, proactive approach to develop guidance. Its understanding of the strengths and weaknesses inherent with RWE and its use in patient care versus regulatory decision-making is evolving quickly,” Qin Ye, M.D. global real-world evidence lead at ZS Associates, told BioSpace.
“What FDA hasn’t yet published is explicit recommendations and encouragement about how to change the status quo,” Ye said. For example, “Electronic health records (EHRs) have a lot of opportunity to improve…particularly around unstructured notes. There is more emphasis on efficiency and less on the quality of data capture within patient care settings. So, while physicians can read between the lines of the EHR information, the FDA can’t be expected to do the same.”
Organizations, including the Friends of Cancer Research, Duke University, the FDA and pharmaceutical companies, are working to bridge such gaps as well as to identify robust real-world endpoints that can be used for decision-making. “RWE data centers around the patient, not necessarily clinical endpoints so, while there is a lot of potential, it requires EHR and health IT industries, healthcare providers and pharmaceutical companies to collaborate closely,” Ye said.
Califf and the FDA recognize that incorporating RWE into clinical trials has the potential to reveal previously unknown synergies that may not only improve patient care but may also make the trials themselves more efficient and productive. Expanding the Agency’s data-driven approach to incorporate more patient-driven evidence has the potential to improve the drug development environment not only for developers but also for patients.
